Re

fe
Pl
r en
ea c
se eM
Do

Network+ Certification
No at
tC er
op
y ial
Network+ Certification

ial
To access your Online ANYTIME Learning Components, please go to
http://onlineanytime.newhorizons.com and use the username and password
provided to you in class.

er
WBT Username:

WBT Password:

at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re
 NETWORK+ CERTIFICATION
Course Number: NH77700 (0751EBEE)
Course Edition: 3.2
For software version: na

ial
ACKNOWLEDGEMENTS
Project Team
Curriculum Developers and Technical Writers : Gail Sandler and Pamela J. Taylor • Copy Editor : Taryn
Chase • Reviewing Editor : Tom Elston • Layout Technicians : Kristi Toscano and Isolina Salgado • Quality

er
Assurance Analyst : Frank Wosnick

Project Support
Managing Editor, Internet & Programming : Carlene Jo Kline • Managing Editor, Personal Computing and
Design : Cheryl Russo • Managing Editor, IT Professional : Susan B. SanFilippo • Manager of Instructor-

at
Led Web Curriculum : Joy Morris • Managers of Web Development : Joseph Blakely and Tracy Dedes •
Manager of Publishing Services : Mark Onisk • Publishing Services Supervisor : Rachel Miller • Publishing
Technical Specialist : Daniel L. Quackenbush • Instructional Design Specialist : Susan L. Reber • Project
Technical Support Specialist : Edward Hyder • Testing & Layout Technician : Bonnie S. Watts

y
Administration

op
eM
Vice President of Content & Operations : Paul A. Krause • Director of Advanced Technology Content : Kent
Michels • Director of Personal Computing and Design Content : Joy Insinna • Director of Interactive
Content : Jeff Felice • Director of Online Education : Lesley Darling • Director of Operations & Publishing
Services : Kress Riley

NOTICES tC
No
DISCLAIMER: While Element K Content LLC takes care to ensure the accuracy and quality of these materials, we cannot guarantee their accuracy, and all materials are
c
provided without any warranty whatsoever, including, but not limited to, the implied warranties of merchantability or fitness for a particular purpose

TRADEMARK NOTICES: Element K Content LLC, ZDU, FirstEdition, Element K Journals and the corresponding logos are trademarks and service marks of Element K
Content LLC. Windows NT and NetWare are registered trademarks of Microsoft, Inc., and Novell, Inc., respecitvely. All other product names and services used throughout this
book are common law or registered trademarks and service marks of their respective companies. Use of another entity’s product name or service in this book is for editorial
en

purposes only. No such use, or the use of any trade name, is intended to convey endorsement or other affiliation with the book.
Do

Copyright © 2000 Element K Content LLC. All rights reserved. This publication, or any part thereof, may not be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopying, recording, storage in an information retrieval system, or otherwise, without express written permission of Element K
Content LLC, 500 Canal View Boulevard, Rochester, NY 14623, (800) 434-3466. Element K Content LLC’s World Wide Web site is located at www.elementkcourseware.com

Unauthorized reproduction or transmission of any part of this book or materials is a violation of federal law. If you believe that this book, related materials, or any other
Element K Content LLC materials are being reproduced or transmitted without permission, please call 1-800-478-7788.
se
r
fe
ea
Pl
Re

ii
NETWORK+ CERTIFICATION CONTENT
OVERVIEW

ial
About This Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Lesson 1: Basic Networking Knowledge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

er
Lesson 2: Servers and Clients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Lesson 3: Introduction to the OSI Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

at
Lesson 4: The Physical Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

y
Lesson 5: The Data Link Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

op
eM
Lesson 6: Data Transmission . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Lesson 7: The Network Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

tC
Lesson 8: The Transport Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
No
Lesson 9: The Session Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
c

Lesson 10: TCP/IP Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223

en
Do

Lesson 11: IP Addressing, Subnetting, and Supernetting . . . . . . . . . . . . . . . . . . . . 259

Lesson 12: Accessing and Using a TCP/IP Network . . . . . . . . . . . . . . . . . . . . . . . 289
se
r

Lesson 13: Remote Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

fe

Lesson 14: Network Security and Fault Tolerance . . . . . . . . . . . . . . . . . . . . . . . . 337

ea

Lesson 15: Preparing for Network Installation . . . . . . . . . . . . . . . . . . . . . . . . . . 381

Pl
Re

Lesson 16: Maintaining the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

Lesson 17: Troubleshooting the Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423

Contents iii
CONTENT Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 443

OVERVIEW Glossary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 525

ial
er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

iv
NETWORK+ CERTIFICATION CONTENTS

ial
About This Course
About This Course . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Course Setup Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii

er
How to Use This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv

LESSON 1: BASIC NETWORKING KNOWLEDGE

at
TOPIC 1A Network and Internetwork Concepts . . . . . . . . . . . . . . . . . . 2

y
TOPIC 1B Network Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

op
TOPIC 1C
eM
Networking Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Lesson 1 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

LESSON 2: SERVERS AND CLIENTS

TOPIC 2A Network Operating System Basics . . . . . . . . . . . . . . . . . . . . 18
tC
No
c

TOPIC 2B Network Clients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Apply Your Knowledge 2-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
en

TOPIC 2C Directory Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Do

Check Your Skills 2-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

Lesson 2 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

LESSON 3: INTRODUCTION TO THE OSI MODEL

se
r
fe

TOPIC 3A Introduction to the OSI Model . . . . . . . . . . . . . . . . . . . . . . . 50

ea

Apply Your Knowledge 3-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

TOPIC 3B Families of Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Pl
Re

Lesson 3 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

Contents v
CONTENTS LESSON 4: THE PHYSICAL MEDIA
TOPIC 4A Data-transmission Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Apply Your Knowledge 4-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

TOPIC 4B

ial
Unbounded Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

TOPIC 4C Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

TOPIC 4D Signal Transmissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

Lesson 4 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

er
LESSON 5: THE DATA LINK LAYER
TOPIC 5A

at
Data Link Layer Considerations . . . . . . . . . . . . . . . . . . . . . .100

TOPIC 5B Media Access Control (MAC) and Logical Link Control

y
(LLC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .101
Check Your Skills 5-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

op
eM
TOPIC 5C IEEE 802 Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110
Check Your Skills 5-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

tC
Lesson 5 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

LESSON 6: DATA TRANSMISSION

No
c

TOPIC 6A Introduction to Network Connection Devices . . . . . . . . . . .130

Check Your Skills 6-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
Apply Your Knowledge 6-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
en
Do

TOPIC 6B Network Adapters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .142

Apply Your Knowledge 6-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Apply Your Knowledge 6-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

TOPIC 6C
se

Putting Data on the Cable . . . . . . . . . . . . . . . . . . . . . . . . . . .154

r

TOPIC 6D Network Topologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159

fe

Apply Your Knowledge 6-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

ea

TOPIC 6E Bridges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169

Check Your Skills 6-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Pl
Re

Apply Your Knowledge 6-5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184

Lesson 6 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185

vi
LESSON 7: THE NETWORK LAYER CONTENTS
TOPIC 7A Addresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .188
Check Your Skills 7-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190

TOPIC 7B

ial
Internetwork Connection Devices . . . . . . . . . . . . . . . . . . . . .190

TOPIC 7C Routers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .193

Apply Your Knowledge 7-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
Lesson 7 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201

er
LESSON 8: THE TRANSPORT LAYER
TOPIC 8A Transport Layer Addressing Issues . . . . . . . . . . . . . . . . . . . .204

at
TOPIC 8B The Transport Layer Segmenting Process . . . . . . . . . . . . . . .207

TOPIC 8C

y
Transport Layer Connection Services . . . . . . . . . . . . . . . . . .209
Apply Your Knowledge 8-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

op
eM
Lesson 8 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212

LESSON 9: THE SESSION LAYER

TOPIC 9A tC
Dialog Control Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .216
No
TOPIC 9B Managing the Session . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .220
c

Lesson 9 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

en

LESSON 10: TCP/IP FUNDAMENTALS

Do

TOPIC 10A What is TCP/IP? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .224

TOPIC 10B The TCP/IP Protocol Suite . . . . . . . . . . . . . . . . . . . . . . . . . . .229

se
r

Apply Your Knowledge 10-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

Lesson 10 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
fe
ea

LESSON 11: IP ADDRESSING, SUBNETTING, AND SUPERNETTING

Pl
Re

TOPIC 11A IP Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .260

TOPIC 11B Interior Gateway Protocols . . . . . . . . . . . . . . . . . . . . . . . . . .268

TOPIC 11C Introduction to Subnetting . . . . . . . . . . . . . . . . . . . . . . . . . .279

Apply Your Knowledge 11-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284

Contents vii
CONTENTS TOPIC 11D Supernetting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .285
Lesson 11 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287

LESSON 12: ACCESSING AND USING A TCP/IP NETWORK

ial
TOPIC 12A Configuring Clients for TCP/IP Networks . . . . . . . . . . . . . . .290
Check Your Skills 12-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293

TOPIC 12B TCP/IP Suite Utilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .294

er
TOPIC 12C Using TCP/IP Suite Utilities to Check Network Connectiv-
ity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .307
Apply Your Knowledge 12-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
Check Your Skills 12-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
Lesson 12 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

at
LESSON 13: REMOTE ACCESS

y
op
TOPIC 13A
eM
Remote Access Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . .318

TOPIC 13B Examining Remote Connection Options . . . . . . . . . . . . . . . .321

TOPIC 13C
tC
Dial-up Networking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .326
Apply Your Knowledge 13-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Check Your Skills 13-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
No
Lesson 13 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
c

LESSON 14: NETWORK SECURITY AND FAULT TOLERANCE

en
Do

TOPIC 14A Security Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .339

TOPIC 14B Password Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .347

TOPIC 14C
se

Data Encryption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .349

r

TOPIC 14D Firewalls and Proxy Servers . . . . . . . . . . . . . . . . . . . . . . . . . .352

fe

Apply Your Knowledge 14-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

ea

TOPIC 14E System Fault Tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . .355

Apply Your Knowledge 14-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
Pl
Re

Apply Your Knowledge 14-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371

TOPIC 14F Backing Up Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .372

Apply Your Knowledge 14-4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
Lesson 14 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380

viii
LESSON 15: PREPARING FOR NETWORK INSTALLATION CONTENTS
TOPIC 15A Planning and Gathering Information . . . . . . . . . . . . . . . . . .382

TOPIC 15B Environmental Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .385

ial
TOPIC 15C Networking Components and Connectors . . . . . . . . . . . . . .389
Apply Your Knowledge 15-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394

TOPIC 15D Compatibility Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .396

Apply Your Knowledge 15-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398

er
Lesson 15 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

LESSON 16: MAINTAINING THE NETWORK

at
TOPIC 16A Network Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . .402

TOPIC 16B

y
Backup Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .405

op
TOPIC 16C
eM
Virus Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .412
Apply Your Knowledge 16-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416

TOPIC 16D Patches and Other Software Updates . . . . . . . . . . . . . . . . . .417

tC
Apply Your Knowledge 16-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
Lesson 16 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
No
LESSON 17: TROUBLESHOOTING THE NETWORK
c

TOPIC 17A Exploring Troubleshooting Models . . . . . . . . . . . . . . . . . . . .424

en
Do

Apply Your Knowledge 17-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434

TOPIC 17B Identifying Troubleshooting Tools . . . . . . . . . . . . . . . . . . . .435

Apply Your Knowledge 17-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
Apply Your Knowledge 17-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
se
r

Lesson 17 Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442

Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .443
fe
ea

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .513

Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .525
Pl
Re

Contents ix
Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
 ABOUT THIS
COURSE

ial
ABOUT THIS COURSE
Network+ Certification is an instruction book that will introduce
you to networking theory in a non-vendor-specific course.

er
Course Prerequisites
We designed Network+ Certification for the student with 18 to
24 months of professional computer support experience. It is assumed that you

at
have a general working knowledge of personal computers, some network experi-
ence, and have taken the following courses (or have equivalent experience):
Windows 95 - Level 1, or Windows 98 - Level 1.

y
op
Course Objectives
eM
When you’re done working your way through this book, you’ll be able to:

• Describe the purpose of networking protocols and networking in general.

•

•
to access them.
Describe the OSI networking model and its relationship to networking
tC
Identify features of various network operating systems and the clients used
No
components.
c

• Describe the network components relating to the Physical layer of the OSI
model.
• Describe the function of the Data Link layer of the OSI model.
en

•
Do

Explain how data is transmitted over a network.

• Describe the function of the Network layer of the OSI model.
• Describe the function of the Transport layer of the OSI model.
• Describe the function of the Session layer of the OSI model.
se

•
r

Explain the fundamental concepts of the TCP/IP protocol suite.

• Explain the use of TCP/IP addresses and subnets.
fe

• Access and use a TCP/IP network.

ea

• Describe the requirements for remote network access.

Pl
Re

About This Course xi

 •
ABOUT THIS Explain the need for and ways to implement network security and fault
tolerance.
• Describe the types of information that need to be gathered prior to installing
COURSE •
or updating a network operating system.
Explain ways to monitor and manage a network.
•

ial
Describe a systematic approach for troubleshooting network problems.

COURSE SETUP INFORMATION

er
Hardware and Software Requirements
While this course can be taught using only Windows 95/98 workstations for each
student and a classroom Windows NT 4 Server, the course is written to also

at
include a NetWare 4.11 server and optionally, a Linux server. Red Hat Linux 6.0
was used to develop this course.
To run this course, you will need:

y
• For each server:

op
—
eM
At least 32 megabytes (MB) of Random Access Memory (RAM)
— 1.2 GB hard drives
— A Pentium or faster processor
—
—
—
A mouse or compatible tracking device
A 3.5″ 1.44 MB floppy-disk drive
tC
A VGA or higher resolution video card and monitor
No
— A CD-ROM drive
c

— Windows NT Server 4.0 (with Service Pack 4 or higher applied)

— NetWare 4.11 (optional)
en

— Linux (optional)
Do

• For each workstation:

— At least 16 megabytes (MB) of Random Access Memory (RAM)
— 500 MB hard drives or larger
se

— A Pentium or faster processor

r

— A VGA or higher resolution video card and monitor

fe

— A mouse or compatible tracking device

ea

— A 3.5″ 1.44 MB floppy-disk drive

— A CD-ROM drive
— Windows 95
Pl
Re

— Internet access
— An email account for each workstation

xii
— Headphones or speakers (If students will be doing the video presentations
included on the Interactive Learning CD each from their own workstation,
issue each workstation a set of headphones. If you’ll be playing the video
ABOUT THIS
presentations and displaying them to students from the Instructor’s worksta-
tion, you’ll need speakers so students can hear the audio portion of the COURSE
presentations.) A sound will also be required to use the video presentations.

ial
— All systems require network cards and all systems are networked together.
(We used 10-Base T networking while developing this course.)

Class Requirements
In order for the class to run properly, perform the procedures described below.

er
Install and configure classroom Windows NT 4 server:
1. Partition and format the hard disk into a primary, DOS-bootable FAT
partition.

at
2. Install DOS 6.2x or higher on the FAT partition in the C:\DOS directory.

3. Create an AUTOEXEC.BAT file that includes at a minimum, the line

y
@path=C:\DOS.

op
4.
eM
Install Windows NT Server 4.0 as the PDC of NETDOMAIN. Name the
computer NTSRVR. Install NetBEUI and NWLink. Install TCP/IP and con-
figure it as required for your classroom. Don’t install Internet Information

tC
Server (IIS). Change the Administrator password to password. Create an
Emergency Repair Disk.

5. Install a printer named class. Share it as class. (Don’t install a print device.)
Pause the printer. Install a printer named class . Share it as class . (Don’t
No
install a print device.)
c

6. Create a user account called Domain_Admin with the password of manage,

uncheck User Must Change Password At Next Logon, and add the account
en

to the Domain Admins group.

Do

7. Create a user account for each student in the class (Stu1 through Stu#).

Install and configure the classroom NetWare 4.11 server:

se

1. Create a new server using the Simple Installation method.

r

2. Name the server 411SRVR.

fe
ea

3. The server will contain only one volume, SYS:.

4. The Context for this server and its objects should be CLASS.
Pl
Re

5. Accept defaults as appropriate.

Install and configure the classroom Linux server:

1. Boot from the Linux CD-ROM, and install using the Server Installation
Class.

About This Course xiii

ABOUT THIS 2.

3.
Accept defaults as appropriate.

Name the server linuxsrvr.

COURSE 4. Assign the password Passw0rd (with a capital P and the number 0) to the
root account.

ial
5. Configure TCP/IP as appropriate to your classroom.

6. Create a user account for each student in the class (user1 through user#).
You can use the adduser command to add the users (log in as root, then
enter adduser user#). After creating the users, use the passwd command to
assign each student a password.

er
Install and configure student and instructor workstations:
1. Format the computer hard drive so that it is running only DOS.

2. Run Windows setup. Select Typical installation. Allow Windows 95 to detect

at
your computer’s hardware. Select Network Adapter, if available. We
assumed the following settings when developing the course:
•

y
The floppy-disk drive is A.
• A Network Neighborhood is present. If you don’t have a Network

op
eM
Neighborhood, you may not be able to perform some of the tasks.

3. Select the option to install the most common Windows 95 components.

tC
4. Set the appropriate time zone.

5. Install the printer driver of your choice (this course was written using the HP
LaserJet 4Si printer driver). Name the printer Class Printer.
No
You don’t need a printer to
c

6. Configure TCP/IP as necessary for your classroom.

run the course, you just need
a printer driver. 7. Install Dial-up Networking. (In the Add/Remove Programs Control Panel
utility, display the Windows Setup page. Select Communications, and then
en

click Details. Select Dial-up Networking, and click OK. Accept defaults as
Do

This configuration eliminates

prompted.)
variances in user settings,
Start-menu groupings, DOS
environmental settings, and 8. Set up the Network+ Interactive Learning CD on each workstation by run-
device-driver conflicts that ning SETUP from the CD-ROM.
might result from upgrading
se
r

a previous version of 9. Install and configure both Client for Microsoft Networks and Novell Client
Windows. software on each workstation. For the Novell Client software, specify the
fe

default server as 411SRVR.

ea

10. Copy the CAB files to C:\CAB on each workstation, or make the Windows
95 CDs available for use if needed during class.
Pl

11. Log in to the NetWare server as Admin. In NetWare Administrator, create an

Re

Organizational Unit container under Class named NetPlus. Create some users
and a group under the NetPlus container.
If you have examples of any of the following which students can see (and possi-
bly examine in more detail), it would enhance students’ learning experience.
Items of special interest might include: NICs, patch panels, cables, repeaters,
hubs, bridges, routers, brouters, print services devices, UPS, and so forth.
xiv
If you have examples of any network media that students could view and possi-
bly manipulate, it would be beneficial to students. This might include: various
connector types such as RJ-45, RJ-11, BNC, IBM genderless, and IBM external
ABOUT THIS
transceivers; vampire taps; cable-making tools; and pieces of cable to create
coaxial and 10BaseT cables. COURSE
One of the optional activities in the course is to ″create a network without

ial
computers.″ This involves students locating a network card and the appropriate
cabling to go with it, any additional devices such as hubs or MSAUs, and attach-
ing them all together. If you have several different topologies that students could
choose from, it would make them think about what pieces really go together.

er
HOW TO USE THIS BOOK
You can use this book as a learning guide, a review tool, and a reference.

at
As a Learning Guide

y
Each lesson covers one broad topic or set of related topics. Lessons are arranged
in order of the OSI model, followed by topics that build on this knowledge. For

op
eM
this reason, you should work through the lessons in sequence.
We organized each lesson into explanatory topics and step-by-step activities. Top-
ics provide the theory you need to prepare for Network+ certification; activities
allow you to apply this theory to practical hands-on examples or for you to dis-

tC
cuss the concepts presented.

As a Review Tool
No
Any method of instruction is only as effective as the time and effort you are will-
c

ing to invest in it. For this reason, we encourage you to spend some time
reviewing the book’s more challenging topics and activities.
en
Do

As a Reference
You can use the Concepts sections in this book as a first source for definitions of
terms, background information on given topics, and summaries of procedures.
se
r
fe
ea
Pl
Re

About This Course xv

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Basic Networking LESSON
Knowledge
1

ial
Data Files:
none
Overview
Lesson Time:

er
This lesson introduces the concept of a network and how the network com- 45 minutes
puters are connected.

at
Objectives
To understand the purpose of the networking protocols and networking in general,

y
you will:

op
1A Define network concepts.
eM
This topic identifies and explains the roles played by basic network
components. It also defines interconnectivity and interoperability in terms

tC
of networking.

1B Differentiate between physical and logical topologies.

Now that you know what a network is, you’ll examine the physical topol-
No
ogy or structure of the network and the logical topology or path of a
c

signal on the network.

1C Define networking strategies.

en

You’ll examine two major types of networks—client-server and peer-to-

Do

peer.
se
r
fe
ea
Pl
Re

Lesson 1: Basic Networking Knowledge 1

 TOPIC 1A
Network and Internetwork Concepts
Network computing results from three basic needs: the need to share information;

ial
the need for a way, or path, of communicating with others; and the need for rules
to control how communication occurs. This course addresses these three basic
network computing needs.
A computer network is a collection of hardware and software that enables a group
of nodes (devices) to communicate and provides users with access to shared
resources. Nodes can be computers, such as mainframe computers, minicomput-

er
network: ers, or personal computers; they can also be devices, such as data-entry terminals.
A collection of hardware and The shared resources to which network computers can provide access include
software that enables a group files, applications, and other software; and hardware, such as printers or modems.
of computers to
communicate and provide
users with access to shared

at
resources. Basic Network Components
A network or other communication system has messages that need to be commu-

y
nicated, transmitters and receivers (senders and recipients of the message), and a
node:

op
communication channel through which the transmitter sends the message to the
A network access point.
eM
receiver. Nodes or stations are computers or other devices that communicate with
Examples include terminals
each other in a computer network. In network communication, nodes typically
and computers.
function as transceivers; they can transmit and receive messages.

message:
A chunk of data that is
transmitted over a network.
tC
No
c

transmitter:
en

The component on the

Do

″speaking″ end of a
transmission.

receiver:
se
r

The component on the

″hearing″ end of a
transmission.
fe
ea

transceiver:
Figure 1-1: Basic network components.
Pl

A device that can function as

Re

a transmitter or receiver.
Let’s look at the fundamental components of computer networks:
• Two or more nodes that desire to share information.

2
• Cabling or some other medium through which network nodes can
communicate. Typically, the network communication channel is cable; how-
ever, network communication channels can consist of other media, such as
satellite, infrared light beams, and radio.
• Software that enables network nodes to communicate. Network nodes must
use the same method or language to communicate. Network software imple-

ial
ments communication protocols. Some computer operating systems (such as
MS-DOS) require that communication software be added for network com-
munication because the systems weren’t originally designed for this type of
communication. Other computer operating systems, such as UNIX, NetWare,
Windows NT, or Windows 95/98, have built-in communication capabilities.

er
Component/Feature Definition
Server Any computer that makes its hardware and software resources available
for use by other computers on the network.
Client Any computer that uses the resources of the server.
Workstation A high-performance personal computer with specialized operating system

at
software to provide features such as multitasking, multiprocessing, and
security. Typically, workstations are used in situations where users need
to run complex and process-intensive applications on their desktop

y
computers. The term ″workstation″ is often used when ″client″ would be
more appropriate.

op
Media
eM
The physical connections between computers, such as network cabling.
Resources Files, printers, or other items accessed by clients.
Shared data Information, typically a file, provided by one computer to another across

tC
the network.
Shared printers and other Any hardware resource provided by a server to a client across a network.
peripherals
No
At the most basic level, a computer network consists of at least two computers
c

connected by a communications link, such as a physical cable. Typically, one of

the computers acts as a server, making its hardware and software resources avail-
able for use by other computers on the network. A computer that uses the
en

resources of the server is called a client .

Do
se
r
fe
ea
Pl
Re

Lesson 1: Basic Networking Knowledge 3

 ial
er
at
y
op
eM
Figure 1-2: A computer network.
The main trunk of a network communication channel is referred to as the network
backbone . Network devices can be attached directly to this backbone or be con-

tC
nected through other network devices. Cables can be connected to this backbone;
these secondary communication channels connected to the backbone are called
backbone: segments or a portion of the backbone itself.
The main trunk of a network
communication channel.
No
TASK 1A-1:
c

segment:
Identifying network components
en

Secondary communication
Do

channel connected to the

backbone. Objective: Choose the best answer for each question.

1. Which of the following correctly identifies a computer network?

a. Four computers that share a printer through a printer-sharing
device.
se
r

b. Two computers connected by a temporary serial-port connection

fe

and using file-transfer software.

ea

c. A group of computers using hardware and software to communicate

and provide users with access to shared resources.
d. Two computers, and their users, that share information through the
Pl
Re

exchange of floppy disks.

2. A server is a computer that:

a. Makes its resources available to other computers on the network.
b. Uses resources available from other computers on the network.

4
 c. Uses resources from, and makes available its resources to, other
computers on the network.

3. What are the fundamental components of a network?

ial
4. Compare the terms backbone and segment.

5. In Figure 1-3, identify the segments and the backbone.

er
at
y
op
eM
tC
No
c
en
Do

Figure 1-3: Identify the segments and the backbone.

Networking Overview
se
r

Network computing enables computers to share information and resources. How-

fe

ever, standards (rules) need to be established so that communication can occur.

ea

For example, if you’re doing volunteer work in a foreign country but don’t speak
the language and the people there don’t speak English, all of you would become
frustrated because you couldn’t understand each other. Luckily, a communication
foundation exists that includes things like interpreters and language translation
Pl
Re

books that enable people to communicate when otherwise they couldn’t. Stan-
dards make life easier for people by helping daily operations run smoothly.
Without them, chaos would result from people doing things any way they wanted,
such as driving down the middle of the street.

Lesson 1: Basic Networking Knowledge 5

 Network computing uses a similar communication foundation, which enables
computers that ″speak different languages″ to exchange data. It consists of proto-
cols , which are the means by which networking can occur. Every protocol
standardizes tasks associated with computer communication. You can think of a
protocol as a set of rules and regulations that govern a specific aspect of data
protocol: communications. A protocol might define, for example, a method for properly
A code or set of rules by beginning and ending communication and a sequence of events that should occur.

ial
which communication is
initiated, maintained, and A protocol suite is a set of related protocols that come from a single developer or
terminated. source.
• Transmission Control Protocol (TCP) is an example of a protocol.
• The Internet Protocol suite refers to a protocol suite that includes TCP and
several other related data-communication protocols.

er
Transmission Control
Protocol (TCP): A protocol stack is a set of two or more protocols that work together, with each
The component of the TCP/ protocol covering a different aspect of data communications. Typically, TCP and
IP protocol suite responsible
for establishing a connection
Internet Protocol (IP) are used in combination and might be called a protocol
between the sender and the stack.

at
receiver in a network
transmission. It guarantees
delivery of the data through
Internetworking Concepts

y
acknowledgments and
timers.
When you establish communication links between two or more networks, you

op
eM
create an internetwork. An internetwork, then, is a network of networks. You
might consider the formula for internetworking to be I=I+I (internetworking
protocol stack: equals interconnectivity plus interoperability).
A layered set of protocols

tC
working together to provide a
set of network functions.
Each protocol layer services
the layer above by using the
No
layer below.
c

Internet Protocol (IP):

en

The component of the TCP/

Do

IP protocol suite that

addresses and routes data
packets across the network
or internetwork.
se
r

internetwork:
fe

A network of networks.
ea
Pl
Re

Figure 1-4: An internetwork is a network of networks.

6
Interconnectivity
Interconnectivity refers to the ease and ability of connecting networks together.
How easily two networks connect depends, among other factors, on the type of
communication channel each network uses, and the method through which nodes
on each network access the communication channel. If both networks use the interconnectivity:
same type of communication channel and the nodes access it the same way, then The ease and ability of

ial
interconnection is relatively simple. The more disparate the systems, the more connecting networks
together.
complex interconnectivity becomes. For example, connecting two NetWare net-
works would be less complex than connecting a Macintosh AppleTalk network
and an IBM SNA network.
Internetworking devices, such as bridges and routers, connect networks or por-

er
tions of networks. These devices are discussed in greater detail later in the
course.

Interoperability

at
Interoperability refers to how communication capability takes place, and which
nodes on internetworks can communicate and share resources, such as files and

y
printers. As with interconnectivity, providing communication is a more complex
task with disparate systems than with similar systems. For example, there would interoperability:

op
How internetworks
eM
probably be a higher degree of interoperability between two IBM personal com-
puters than between an IBM personal computer and a Macintosh computer. communicate and share
resources.
Although it is an oversimplification, you might find it helpful to think of
interoperability as a software compatibility issue and interconnectivity as a hard-

tC
ware issue.

Gateways
No
c

A gateway is software or hardware (or a combination of the two) that intercon-

nects disparate types of networks, translating as necessary between the two. For
example, a gateway might connect PCs on a LAN to a mainframe. The gateway
en

is often a system dedicated specifically to the gateway task because it requires a gateway:
Do

lot of resources. Software or hardware that

connects disparate types of
Some common gateway types are described in the following table. networks, translating as
necessary between the two.
Gateway type Connects networks that use different
se
r

Protocol Protocols
Format Encoding schemes (for example, ASCII and EBCDIC)
fe

Address File systems and directory structures

ea
Pl
Re

Lesson 1: Basic Networking Knowledge 7

 TASK 1A-2:
Identifying internetworking obstacles
Objective: To identify obstacles to interconnectivity and interoperability.

ial
1. Identify factors that would make it difficult to interconnect two
networks.

er
at
2. Identify factors that would make it difficult for network computers to
share files, printers, and other resources.

y
op
eM
tC
No
c

Types of Networks
Networks come in many shapes and sizes. However, in most cases you can cat-
en

egorize them based on their geographical reach. You’ll likely see three terms
Do

describing the general size of networks: local area network (LAN), metropolitan
area network (MAN), and wide area network (WAN).
A local area network (LAN) is a collection of computers cabled together to form
a network in a small geographic area (usually within one building). A LAN can
be only as large as the physical limitations its cabling will allow. Different
se
r

LAN (Local Area cabling schemes have different limitations (which of course you can get around).
Network): However, the distinction remains—LANs are limited in size. They are typically
fe

A collection of computers inexpensive to set up, run and maintain.

cabled together to form a
ea

network in a small
geographic area (usually
within a building).
Pl
Re

8
A metropolitan area network (MAN) is slightly larger than a LAN and is con-
tained within a single city or metropolitan area. A large company might create a
MAN to link separate office buildings spread throughout a city. MANs are typi-
cally owned by private providers. They are more expensive to implement and MAN (Metropolitan Area
maintain than LANs. Network):
A network linking several
A wide area network (WAN) is relatively large and can span countries and buildings within a city or

ial
continents. Typically, WANs are privately paid for and maintained; however, com- metropolitan area.
panies often share WAN lines with other companies to save money. A company
might purchase part of the capacity of a network connection from a national pro-
vider, like MCI WorldCom. The remaining capacity of that connection will be
sold to other companies who need similar communications capabilities. Equip- Another type of network you
might see is CAN (Campus
ment owned by the network provider divides the network capacity, keeping each Area Network).

er
customer’s data separate and safe. The advantage of a WAN is that it is essen-
tially unlimited in geographic area. However, costs vary widely depending on
how the WAN is set up and configured.
WAN (Wide Area
Network):

at
A network spanning large
TASK 1A-3: areas such as multiple cities,
states, or countries.
Evaluating network basics

y
Objective: Choose the best answer for each question.

op
eM
1. Explain why the following examples are (or are not) a LAN.

A. A large office building contains 487 computers that are cabled

tC
together and share data, applications, and hardware.
No
B. A Utah-based branch office has five computers that share data and a
c

mail application with two computers in an office in Seattle.

en
Do

2. Identify the network type (LAN, MAN, or WAN) that would be most suit-
able for the needs described below. Explain why you chose that type of
network.

A. You need to exchange files and email with other employees in your
se
r

company who are located in offices throughout the United States.

fe
ea

B. You need to share network applications and printers with everyone in

your department; all of the desks are within 100 feet of each other.
Pl
Re

Lesson 1: Basic Networking Knowledge 9

 C. You need to exchange files with other departments in your company.
All departments are in different buildings throughout the city and sur-
rounding suburbs.

ial
TOPIC 1B
Network Topologies

er
A network topology refers to the layout of the transmission medium and devices
on a network. There are two types of topologies: physical and logical. Topologies
use either a point-to-point or multipoint connection scheme. A connection scheme
point-to-point connection indicates how many devices are connected to a transmission media segment or an
scheme: individual cable. An example of a point-to-point connection is a printer or modem

at
Two devices are connected connected to your computer. Another example is two computers connected
by a single communication
channel.
directly to each other to use file-transfer software such as Fastwire, MacLink
Plus, or the Microsoft Windows direct cable connection. An example of a

y
multipoint connection is a star, bus, or cellular topology network.

op
multipoint connection
eM
scheme:
Connection of three or more
Physical Topology
devices by a communication Each network has a connection scheme that describes, in general terms, the trans-

tC
channel.
mission media layout. Each network also subscribes to at least one topology,
which describes the network’s entire physical structure.
In networks that use media cable, the topology defines the cable’s actual physical
network topology:
No
configuration. Physical topologies include:
The layout of the
•
c

transmission medium and Mesh. Provides each device with a point-to-point connection to every other
devices on a network. device in the network.
• Star. Branches out via drop cables from a central hub (also called a multiport
en

repeater or concentrator) to each workstation. A signal is transmitted from a

Do

workstation up the drop cable to the hub. The hub then transmits the signal
physical topology:
to other networked workstations.
A network’s entire physical
configuration. • Ring. Connects workstations in a continuous loop. Workstations relay signals
around the loop in round-robin fashion.
•
se

Bus. Uses a single main bus cable, sometimes called a backbone, to transmit
r

data. Workstations and other network devices tap directly into the backbone
by using drop cables that are connected to the backbone.
fe
ea
Pl
Re

10
• Cellular. Refers to a geographic area, divided into cells, combining a wire-
less structure with point-to-point and multipoint design for device
attachment. The devices in each cell directly communication with a central
station called a hub.
• Hybrid. Combines two or more different topologies.

ial
er
at
y
op
eM
Figure 1-5: Physical topologies.

tC
No
Logical Topology
c

The network path that a signal follows is called the logical topology. Two logical
topologies are frequently used.
•
en

Ring. Generates and sends the signal on a one-way path, usually

Do

counterclockwise.
logical topology:
• Bus. Generates and sends the signal to all network devices. The network path through
which data travels.

TASK 1B-1:
se
r

When referred to as a logical

topology, ring and bus
Reviewing the Physical and Logical topologies
fe

topologies usually have the

word ″logical″ before them;
ea

Objective: Using the following clues, identify the appropriate topology. for example logical ring
topology.
1. Covers a geographical area.
Pl
Re

2. Uses point-to-point connection to every network device.

Lesson 1: Basic Networking Knowledge 11

 3. Workstations connect to the network in a continuous loop.

4. Uses drop cables to connect to a backbone.

ial
5. Uses drop cables to connect to a central hub.

6. Uses multiple topologies.

er
7. Generates and sends signals on a one-way path.

at
8. Generates and sends the signal to all network devices.

y
9. The structure or layout of a network’s transmission media.

op
eM
10. The path a signal follows through a network.

TOPIC 1C tC
No
c

Networking Strategies
Networks are characterized by the types of strategies they employ to interconnect
en

computers. Let’s examine the following strategies:

Do

• Types of node interaction;

• Processing strategies; and
• The services provided by networks.
se
r

Node-connecting Strategies
fe
ea

Three types of relationships exist among network nodes: hierarchical (or legacy),
client-server, and peer-to-peer.
Pl
Re

12
Hierarchical Networks
Some computer systems (typically those based on mainframes and minicomput-
ers) provide a host-to-terminal (also called primary-to-secondary) relationship hierarchical network:
between nodes. These networks are also referred to as legacy networks . Some- A network in which a host
times, you might need to add your network to an existing legacy network. This controls network
type of network can be hierarchical or centralized. In central processing, the host communications and

ial
or server performs all of the processing; users interact with the host through processing.
dumb terminals that have no processing power.
Some advantages of this type of network include centralized administration and
inexpensive terminals or nodes. Disadvantages include expensive site preparation
for the server, lack of off-the-shelf applications, processing power that diminishes

er
with each additional user login, and the expense of purchasing and upgrading
these networks.

at
y
op
eM
Figure 1-6: A hierarchical network.
tC
No
c

Client-server Networks
Servers are computers that perform a service on behalf of other network devices.
en

There are several types of servers. For example, a computer that provides other client-server network:
Do

network nodes with access to network storage devices is a file server . Print serv- A network in which duties
ers provide other network nodes with access to a printer. Both Microsoft are divided between clients,
Windows NT Server and Novell NetWare network operating systems use the which are users of network
client-server model. services, and servers, which
are providers of network
This type of network typically uses distributed processing. Applications designed services.
se
r

for client-server networks are typically stored on network drives. When a user
runs the application from a workstation, the application is loaded into the work-
fe

station’s memory, where it runs much like a non-network (local) application. This
scheme is called distributed processing because the processing of applications
ea

occurs at client stations, not in the server.

Advantages of client-server networks include processing power can be localized
to point of need, overall cost is usually much less than a hierarchical network,
Pl
Re

and servers can be optimized to provide particular services. Some disadvantages

include less centralized administration and more expensive nodes than the dumb
terminals used in legacy networks.

Lesson 1: Basic Networking Knowledge 13

 ial
er
at
y
op
eM
Figure 1-7: A client-server network.

Peer-to-peer Networks

tC
Computers that perform similar functions, both requesting and providing network
services, are peers . Networks in which no single computer controls network
communication functions are called peer-to-peer networks , or direct-cable con-
nections . Another name for this type of network is a workgroup solution .
No
peer-to-peer network: Windows 95/98 and NT Workstation include peer-to-peer support built into the
c

Each workstation on the operating system. UNIX and many of the derivatives such as Linux can operate
network acts as both a server in peer-to-peer networks.
and a client by sharing its
resources with other Advantages include that they are relatively inexpensive, processing power is
en

computers and acessing localized to the point of need (processing happens at the workstation), and all
Do

others’ shared resources. resources are potentially available over the network from every and to every
machine on the network. Some disadvantages are that there is no central adminis-
tration and that security can be difficult to control. Also, when you have more
than 15 nodes on such a network, it doesn’t work well.
se
r
fe
ea
Pl
Re

14
 ial
er
at
Figure 1-8: A peer-to-peer network.

y
Mixed Relationships

op
eM
It is possible for a network to support combinations of relationships. For
example, a network might support hierarchical, client-server, and peer-to-peer
access. In such an environment, a user might be able to access files and other
shared resources from another workstation, from a file server, or from a host.

tC
UNIX systems can be used in any or all of these relationships.
While this provides you with great flexibility and provides access to data from a
wide variety of sources, there can be many disadvantages. Administration can be
complex and the hardware and software to connect these disparate systems can be
No
expensive and complex to install. Off-the shelf solutions are often unavailable, so
c

you often need to mix and match products from several vendors, and possibly
create some of your own solutions. Some vendors won’t support their products
when you use them in such a mixed network.
en
Do

TASK 1C-1:
Identifying node interaction strategies
se
r

Objective: Identify the type of node interaction strategy that best

describes each of the following situations.
fe
ea

1. You can take advantage of several different network types.

Pl
Re

2. A mainframe computer initiates and manages network communication.

3. No single computer controls network communications.

Lesson 1: Basic Networking Knowledge 15

 4. A single computer provides specific network services to clients.

5. You can access information on the Internet.

ial
Summary

er
In this lesson, you’ve defined what a network is, identified physical and
logical topologies, and examined client-server and peer-to-peer networking
strategies.

at
LESSON 1 REVIEW

y
1A What are the three components required to have a computer network?

op
eM
tC
1B Define physical topology and logical topology.
No
c

1C What are the three relationships that exist between network nodes?
en
Do
se
r
fe
ea
Pl
Re

16
Servers and Clients LESSON
2

ial
Overview Data Files:
A network needs clients and servers in order to function. In this lesson, none
you’ll identify features of several network operating systems and the clients
Lesson Time:

er
used to access those operating systems.
1 hour

Objectives

at
To understand servers and clients, you will:

2A

y
Identify the basics of three network operating systems.
You’ll be introduced to three of the most popular network operating sys-

op
eM
tems in this topic: Microsoft Windows NT, Novell NetWare, and UNIX.

2B Identify features of various network clients.

tC
You’ll examine how to connect DOS, Windows, and Macintosh clients to
your network by using built-in and additional utilities.

2C Identify the directory services used in NT and NetWare networks.

No
You’ll be introduced to the domain model used in Windows NT Directory
Services and the Novell Directory Services used in NetWare networks.
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 17

 TOPIC 2A
Network Operating System Basics
The term network services suggests that there is an entity that requests a service

ial
and another entity that provides the service. A service provider consists of a com-
puter as well as the software enabling the computer to fulfill requests. The
network operating system (NOS) contains the applications that provide network
services. While the local operating system controls the resources for the computer
on which it resides, the network operating system manages the network resources
so that they can be adequately shared among the network users.

er
Let’s examine the role of servers, clients, and peers in networks. Servers are gen-
erally service providers, clients are generally service requesters, and peers can be
both. The active software determines whether a computer is a server, client, or
network operating peer. It is possible for one computer to act as a server, client, and a peer at the
system: same time; however, most computers fulfill only one role at a time.

at
A collection of software that
enables network nodes to To work effectively, individuals need access to data, processing, and input/output
communicate with other resources. Through the network operating system, users gain access to network

y
nodes. services. Granting this access to users is transparent.
Network services include file, message, print, security, routing, network adminis-

op
eM
tration, application, and database. These are common to all major network
operating systems.
Let’s examine three different network operating systems. These are Windows NT,

tC
Novell NetWare, and UNIX. They are the three most popular network operating
systems.
No
Windows NT
c

Windows NT is a 32-bit operating system that can run on Intel (and compatible)
and RISC platforms. It is a preemptive multitasking operating system that shares
en

processor time between all running processes. NT is compliant with the govern-
Do

ment’s C2 security specifications. It can be used to build a network, intranet, or

Internet server. Win32 API, 16-bit Windows, and MS-DOS based applications are
supported on the NT platform. NT comes with the Internet Explorer Web browser
and the email software Microsoft Exchange Client.
se
r
fe
ea
Pl
Re

18
 ial
er
at
y
op
Figure 2-1: The Microsoft networking family.
eM
The following table describes four different editions of Windows NT.

tC
Windows NT edition Designed for
Workstation Desktop operating system for business applications.
Server General network operating system for any size network.
No
Enterprise Server Large, mission-critical applications.
c

Terminal Server Non-PC devices, such as dumb terminals, to be part of an NT

network via terminal emulation.
en
Do

The BackOffice runs on NT Server. BackOffice includes:

BackOffice component Is used for

Exchange Server Messaging and collaboration services
Proxy Server Firewall and Web-caching services
se
r

Internet Information Server (IIS) Internet and intranet services

Site Server Intranet, Internet, and e-commerce management services
fe

Systems Management Services Inventory, network diagnostics, and software distribution services
ea

(SMS)
SNA Server Integrating legacy systems with an NT network
SQL Server Relational database management
Pl
Re

NT Server is a network operating system. It includes Gateway Service for inte-

grating NetWare and NT networks and NetWare Migration utility to migrate
NetWare users and data from NetWare servers to NT servers. You can also pur-
chase additional products that enable you to administer both servers from a single
location.

Lesson 2: Servers and Clients 19

 Fault tolerance features of NT Server include disk mirroring, disk duplexing, and
disk striping with parity.
For Internet support, NT Server includes TCP/IP support, Domain Name System
(DNS) server support, IIS, and FrontPage (HTML page editor) software.

ial
er
at
y
op
eM
Figure 2-2: A domain.
tC
NT Server uses a domain model. A domain is a logical grouping of computers
No
and devices on a network administered as a single entity by using a common set
c

of rules and procedures. There can be one or more domains per network.
domain: Domains can be configured to share resources with other domains, but also might
An organizational unit on a not be configured to share resources. Systems identified as being a member of a
en

Windows NT network in domain might be primary and backup domain controllers, member servers, or
Do

which workstations and

servers share a single
client workstations. Users who log in to the domain have access to all servers in
accounts database and are the domain.
administered as a group. NT Servers are generally installed as primary or backup domain controllers. In
order to change domain membership you’d have to reinstall NT.
se
r

Only one primary domain controller exists per domain. It contains the domain
accounts database and the user and group database. It is recommended that at
least one backup domain controller be installed per domain. It contains a copy of
fe

the domain accounts database. Changes made to the primary domain controller
ea

copy of the database are replicated to all backup domain controllers in the
domain. Backup domain controllers can be used to validate user accounts and can
be promoted to a primary domain controller if something happens to the original
Pl
Re

primary domain controller. Geographically dispersed domains should have at least

one backup domain controller at each location so that logons can be handled
locally, reducing traffic over the WAN link.

20
Comparing Domains and Workgroups
NT can also use a workgroup model rather than domains. A workgroup is used on
a peer-to-peer network. It’s typically used in a very small network that doesn’t
require as secure of an environment as that provided by domains. Every computer
in the workgroup maintains its own account information, and doesn’t share it
with other systems. Resources are accessed by knowing and entering the pass-

ial
word assigned to the resource.

er
at
y
op
eM
Figure 2-3: A workgroup.
tC
No
c

NT Superuser
en

Network operating systems need to have a user with full access and rights to
Do

everything on the network in order to administer it. This user is referred to as the
superuser. Different NOSes have different superuser account names.
You can also define additional users with the same rights and access as the
default superuser. This is useful in case you forget the password of the superuser superuser:
se

or if you need to share administration responsibilities between several people.

r

A network account with full

access and rights to
The superuser on an NT system is the user Administrator. This account has com- everything on the network in
fe

plete, unlimited access over the operating system, whether it is used as a order to administer it.
ea

workstation or a network server.

Pl
Re

Administrator:
The superuser account on an
NT system.

Lesson 2: Servers and Clients 21

 NetWare
Novell’s NetWare product is another popular PC network operating system.
Workstations use Novell Client software to log in to and communicate with the
network. NetWare workstation software also enables workstations to determine
Novell was founded in 1983. the operations to be performed by the workstation and the ones to be delegated to
the network. Servers use NetWare to provide and control workstations’ access to

ial
network resources, including shared printers, storage, and communications
devices. NetWare 5.x is the most recent version of the operating system available.
As a network operating system, NetWare provides services to network users. The
following table describes some of these services.

er
Type of service Description
File-storage services Enable users to store files on and retrieve files from the server’s
hard disk.
Novell Directory Services (NDS) Provide administrators and users with a way to manage and
access network resources.

at
Security services Ensure that network data and resources are available only to those
users who should have access to them. NetWare 5.0 has the
government’s C2 security rating.

y
Print services Enable users to share printers that are connected to the network.

op
Application services Enable administrators to set up and configure network-based
eM applications so that network users have easy access to them.
Storage Management Services Enable administrators to back up and restore data located on both
(SMS) servers and workstations. Data is usually backed up to a tape

tC
storage device.
Connectivity services Enable administrators to ensure Internet and remote access to
users who require such access.
Web services Enable administrators to integrate Web servers and solutions on
No
the network.
c

Database services Enable administrators to deliver database solutions on the

network.
Communications services Enables use of third-party communications and messaging
en

systems including GroupWise, cc:Mail, MHS, and SMTP.

Do

Multiprotocol Routing (MPR) Enables you to connect your network to public and private
networks using services such as T1, frame relay, or ISDN.
Symmetric Multiprocessing (SMP) Enables use of SMP utilities on the network.
se
r
fe
ea
Pl
Re

22
NetWare versions 4.0 and higher use a directory service based network, Novell
Directory Services or NDS. NDS maintains information about all network
resources (such as servers, volumes, users, groups, and printers) in a hierarchical,
tree-structured database that resembles an organizational chart. With it, system
administrators can organize network resources according to the way a given NDS:
workgroup is structured—an efficient method for giving users access to the net- Novell Directory Services.
work according to the role they play in the organization. When a user requests a Formerly known as NetWare

ial
Directory Services. An
resource, NDS locates the resource, verifies that the client is valid, and makes the administrative database that
resource available to the client. defines network entities,
NetWare 3.x and previous versions used a bindery-based database to maintain such as users, groups,
servers, and so on. NDS was
information about users and groups. This wasn’t an enterprise-wide database; a introduced in NetWare 4.0,
separate database was required on each server in the network. This means that if and controls access to

er
you need access to three servers, you’ll need three separate accounts (one on each network information,
server). resources, and services. It
allows users to access all
The default protocol for NetWare 5.0 is IP; previous versions used IPX/SPX as network resources with a
the default protocol. Both protocols are supported. single login procedure.

at
NetWare 5.0 includes several additional products with the operating system.
These include a 5-user version of Oracle, Netscape FastTrack Server, FTP Server,
IPX/IP Gateway, DHCP Server, DNS Server and Client, Netscape Navigator, and The default protocol for

y
others. NetWare 4.x and earlier is
IPX/SPX.

op
eM
NetWare Superusers
The superuser on a NetWare system is the user Admin. This account has com-

tC
plete, unlimited access over the operating system. This user is added when the
first server in the tree is installed. You can add other users with full administra-
tive rights, but this is the default superuser for NDS-based NetWare networks. admin:
NetWare 3.x and previous versions used SUPERVISOR rather than Admin, and a The superuser on NetWare
4.x and higher networks.
No
separate SUPERVISOR account was required for each server on the network.
This user has full
c

administrative rights to the

Directory tree objects and to
UNIX the file system.
en
Do

UNIX is a family of operating systems developed at Bell Laboratories in the late

1960s. Because UNIX was designed for software development, the needs of the
beginning user were ignored in favor of speed and accuracy. However, UNIX has SUPERVISOR:
evolved with an ever-increasing focus on the end-user’s needs. Today, many non- The superuser account on
NetWare 3.x and lower
programmers use UNIX effectively in a variety of ways.
networks. This user has full
se
r

administrative rights to
create users, groups, print
services, and to administer
UNIX Features and Benefits
fe

the file system.

ea

The many features and benefits of UNIX include portability, portable applications
software, multiuser operations, multitasking, utilities, pipes, and software develop-
ment tools.
Pl
Re

One of the main reasons for the longevity and popularity of UNIX is its
portability. UNIX runs on almost every type of computer from micros to super
computers. Unlike most other operating systems, UNIX is almost entirely written
in C, a high-level programming language. Therefore, moving UNIX to a new sys-
tem can be done in a relatively short amount of time.

Lesson 2: Servers and Clients 23

 The applications software written for UNIX is also portable. With only minor
modifications, the same UNIX-based applications can run on any computer that
runs UNIX.
UNIX can function as a workstation and as a network operating system. It is
often configured as a server, but can also be part of a peer-to-peer network, or it
can be used as a workstation on a UNIX or other type of client-server or hierar-

ial
chical network.
This multiuser, multitasking operating system can perform multiple tasks for each
user simultaneously. One of the initial objectives of UNIX was to run the same
operating system on different computers and to enable data communications
between them. Consequently, UNIX was designed with networking in mind.
Many companies have changed to UNIX because of its networking capabilities.

er
There are many built-in and third-party applications and utilities that enable
UNIX to be connected to other networks, such as VMS, MVS, NetWare, and NT.
UNIX provides utilities for everything from email to compilers. There are over
300 utility programs and commands in the UNIX environment. The commands

at
are case-sensitive and the command names can be somewhat cryptic. You can use
pipes, one of the most famous UNIX features, to combine several simple utilities
to perform more complex functions.

y
There are many versions of UNIX available. Some of the major UNIX vendors
include Pyramid, HP, IBM, and Sequent.

op
eM
There are also a lot of UNIX-compatible operating systems, such as Linux. Linux
is a POSIX based operating system that can run on many hardware platforms
including PC, Macintosh, Sparc, and Alpha systems. Some vendors that supply

tC
Linux was developed by Linux are Red Hat, Caldera, Suse, and Debian.
Linus Torvolds at the
University of Finland. It is a
freely available operating
system with open source
No
code.
c

POSIX:
en

Portable Operating System

Do

Interface for UNIX. A set of

standards developed by IEEE,
specifying components of a
portable, vendor-independent
operating system. The goal
of POSIX is to enable the
se
r

development of portable
application software across
fe

different vendors’ hardware.

ea
Pl
Re

24
UNIX Superuser
The superuser on a UNIX system is the user root. This account has complete,
unlimited access over the operating system, whether it is used as a workstation or
a network server.
root:
The superuser on UNIX

ial
systems. This is a special
Cross-platform Communication login name used by the
system administrator to gain
Since many networks today combine two or more different network operating full administrative access to
systems into a single network, you might need a solution in which all servers, as the system. Logging in as
well as clients, can communicate with one another. Some of the solutions in Win- root enables you to perform
dows NT Server that allow NT clients to access NetWare servers include any valid command on any

er
Gateway Service for NetWare, Client Service for NetWare, File and Print Ser- file, regardless of what the
permissions on that file are.
vices for NetWare, and Directory Service Manager for NetWare. NT Server
includes UNIX printing support services.
Novell has a product, NDS for NT, which allows NT users to access both NT and

at
NetWare through NDS with a single login. Novell NOSes also include File and
Print Services for UNIX-based clients. You can also purchase products such as
NetWare for SAA for connecting to IBM SAA mainframes.

y
op
TASK 2A-1:
eM
Identifying features of three popular network

tC
operating systems
1. What are the three major network operating systems?
No
c

2. Who is the administrative user for each of the three major operating
systems?
en
Do

3. What are the core services provided by the network operating systems?
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 25

 TOPIC 2B
Network Clients
Clients are the portion of the network through which users access the network.

ial
They provide the connectivity between the server and the user. The choice of
which client you use can greatly affect the network performance in a client-server
network. Clients can include MS-DOS, Windows (in all of its many varieties
from Windows 3.x to Windows NT Workstations), Macintosh, and NetWare
clients.
Whenever possible, you should choose a client whose operating system is well-

er
matched to the network operating system you’re using. It should be able to run
the protocols available on your server.
You might need to install network-specific software or redirection software on
clients in order for them to communicate with the network. Your client might

at
need to run more than one piece of client software if it needs to connect to vari-
ous network operating systems simultaneously.

y
MS-DOS Clients

op
eM
DOS, by itself, doesn’t include any network connectivity features. Client software
that is DOS-compatible must be installed in order for a DOS-based client to
access the network. The client software that you install will depend on the NOS

tC
to which you are connecting.
If you are using a real-mode client, only one client can be loaded at a time. This
can be a problem if users need to connect to two different network operating sys-
tems, each requiring different client software.
No
real mode:
c

Program instructions can

access any address space in
the first MB of RAM. Usually Windows 3.x and Windows for Workgroups 3.x Clients
en

these programs run in real

While Windows 3.x is network aware, it is not really an operating system since it
Do

mode because they need to

access and use system requires DOS in order to run; therefore, some of the limitations of the DOS client
information. Compare with also apply to these clients.
protected mode.
Many external clients are available for Windows 3.x even though none came with
the software. For most client software, SHARE.EXE should be loaded before
se

loading Windows to prevent multiple applications from corrupting a data file by

r

SHARE doesn’t need to be trying to simultaneously access it.

loaded when using the
fe

Microsoft Network Client Only one network client can be active at once since Windows 3.x uses only real
ea

software. mode network client software. Some examples of the client software available
include:
• Microsoft Network Client
Pl

•
Re

Novell NetWare
• Artisoft LANtastic
• Banyan VINES

26
Each of these clients is used to attach to its own network (for example, NetWare
client attaches the client to a NetWare network, and LANtastic attaches the client
to a LANtastic network).
Windows clients (from
Windows for Workgroups 3.x allows for the use of protected mode software, Windows 3.x through
enabling you to have more than one client loaded at once. It is also a peer-to-peer Windows NT) can all use IPX
networking version of Windows 3.x that enables any system to function as a cli- (called NWLink in Microsoft

ial
ent or a server on the network. terms), TCP/IP, and NetBEUI
protocols to communicate
across the network.

Windows 95/98 Clients

Windows 95 and Windows 98 support both real-mode and protected-mode client protected mode:

er
software. The client software is installed, configured, managed, or removed Program instructions can
through the Network Control Panel. You can only install and use one real-mode access only specific
client at a time. However, you can use as many protected-mode clients at the contiguous areas of the first
same time as you need to use. 640 KB of RAM. The local
operating system typically
Windows 95 real-mode client software includes Banyan VINES DOS/Windows runs in protected mode to

at
3.1, FTP Software NFS client, Novell NetWare Workstation Shell 3.x (NETX), prevent data from being
Novell NetWare Workstation Shell 4.0 and above (VLM), and SunSoft PC-NFS overwritten by other
programs. Compare with real
(5.0). The protected-mode client software includes Microsoft’s Client for NetWare

y
mode.
Networks and Client for Microsoft Networks. The protected-mode clients use no
conventional memory.

op
eM
Windows 98 also includes the Microsoft Family Logon protected-mode client
software. In addition, third-party protected-mode client software is available for
Artisoft LANtastic 7.0 or above, Banyan 32-bit or 16-bit, Digital PATHWORKS

tC
32, and IBM Networks Client for Windows 95.
Novell also has a network client that you can install, called Novell Client (previ-
ously known as Client 32). It is a protected-mode application available for
accessing all versions of NetWare from 2.x through 5.x by using Windows or
No
DOS clients. You can try both Novell Client and the Microsoft Client for
c

NetWare networks to determine which one gives you the best performance on
your network.
en
Do

Macintosh Clients
Macintosh computers come preinstalled with AppleTalk network software. This
enables Macs to be quickly networked to one another; however, if you want to
add your Mac to a NetWare or NT network, you’ll need to install support on the
se
r

client, the server, or both.

NT Server includes Mac client support, but the Services for Macintosh need to be
fe

installed on the server to activate this support. This provides file and printer shar-
ea

ing, and allows you to administer Macintosh users the same way PC users are
administered.
NetWare 4.1x networks are automatically configured with the Macintosh name
Pl
Re

space on the SYS volume to support Macintosh file formats. On other NetWare
versions of NetWare, you will need to load the MAC name space. Also, you need
to install the NetWare Client for Macintosh on each Mac that will be accessing NetWare 5 doesn’t include
the NetWare server. This adds features to the Mac including a login menu on the Macintosh connectivity
support; Prosoft Engineering
menu bar, full file and print resource access, NetWare Volume Mounter utility for
now provides it.
fast NetWare volume access, MacIPX so that Macs can use IPX/SPX, NetWare/IP
support, and support for the built-in Ethernet port on Power Macs.

Lesson 2: Servers and Clients 27

 Linux Clients
Linux workstations are also becoming popular client platforms. Linux supports
TCP/IP, IPX, and NetBEUI protocols, so can communicate with many different
network operating systems.

ial
TASK 2B-1:
Identifying client features
1. Identify which systems in your classroom are clients.

er
2. Your network includes Windows NT 4.0, NetWare 4.11, and Linux 6.0

at
servers. What software would need to be installed on your client in
order to communicate with the network? Is that software included with
the client operating system or does it come from the network operating

y
system?

op
eM
3. What other clients could have been included on this network? What

tC
software would be required for the clients to communicate with the
classroom servers?
No
c
en
Do

Suggested time:
APPLY YOUR KNOWLEDGE 2-1
10 minutes
Logging on to the network and accessing resources
se
r

In this activity, you’ll log on to the network as the NT and NetWare superuser
accounts, then run applications from the servers.
fe
ea

1. If your computer is already at the Windows 95 desktop, choose Start→

Shutdown→Close All Programs And Log On As A Different User.

2. When prompted, log in to the NetWare server as Admin, using the password
Pl
Re

manage. This is the NetWare superuser.

3. When prompted, log on to the NT server as Administrator using the pass-

word password. This is the NT superuser.

28
4. Right-click on Network Neighborhood and choose Map Network Drive. In
the Path text box, enter \\NTSRVR\C$ . This is the default share volume on
an NT Server. Click OK.

5. Open Explorer. Examine the list of drives. The local drives are listed, then
the drive on the NT server you just mapped, followed by the default drive
mappings from the NetWare login.

ial
6. Expand C$ on NTSRVR, Program Files, Windows NT, Accessories, and
open the ImageVue folder. Double-click on WANGIMG to start the Imaging
For NT program from the NT server. You are now running this program
from the server. Choose File→Exit to close the application.

er
7. Expand SYS on 411SRVR, Public, and then open the Win95 folder. Double-
click on NWADMN95 to start the NetWare Administrator utility. You are
now running this program from the NetWare server. Choose Object→Exit to
close NetWare Administrator.

at
TOPIC 2C

y
Directory Services

op
eM
Today, many organizations seek to connect all of their computing resources in a
single, enterprise-wide network—an objective that greatly increases the complex-
ity of system administration. In an enterprise that includes thousands of network

tC
users, even seemingly straightforward tasks (such as creating user names and
authorizing access to network resources) can be troublesome.
Directory services help alleviate some of the problems inherent in administering
large networks. Its main purpose is to regulate access to network information,
No
resources, and services. It is very much like using an operator or the Yellow
c

Pages to find a telephone number or information about a company (or person).

When you are designing your network, you should keep two important goals in
en

mind:
Do

• Each user should be able to access all network resources with a single
account and password.
• You should be able to centrally manage the network from a single computer.
se
r

Windows NT Directory Services

fe

By using Windows NT Directory Services or NTDS, you can create a user only
ea

once in the accounts database, yet you can give this user access to all network
resources. NTDS integrates with Microsoft BackOffice and NetWare; thus, all
users can access these resources through their Windows NT accounts.
Pl
Re

Using NTDS simplifies network management in two ways. NTDS supports cen-
tralized management of the accounts database, no matter how large, from any
computer on your network. NTDS also enables you to define a user’s account
once for your entire network.

Lesson 2: Servers and Clients 29

 When designing your network, you implement NTDS through the use of domains
and trust relationships. You create a Windows NT domain when you group com-
puters and users together for administrative purposes. You add users to the
domain by creating them within User Manager for Domains. You add computers
to the domain either by adding them in Server Manager on a domain controller,
or at the time they ″join″ the domain through their own Control Panel in Win-
dows NT.

ial
Your Windows NT network can consist of one or more domains. Because
domains are used to logically group users and computers, you may find that you
need more than one domain to manage your network. Some of the factors that
influence the domain structure for your network include:
• Geographical —where you establish domains at each geographical location

er
of your company.
• Functional —where you define domains following the company’s organiza-
tional chart (usually departmental).
• Administrative —where you create domains based on who will be managing

at
them.
Once you have designed your domain structure, you might face a dilemma if you
have created multiple domains: what if users need to access resources outside

y
their own domains? Windows NT Directory Services solves this dilemma through
The example in the text is of the use of trust relationships. Through a trust relationship, you can establish a

op
two one-way trusts between
eM
secure connection between two domains. After you create the trust, you can then
domains. The trust
give the users from both domains access to either domain’s local resources.
relationship could also be
one-way only. The goal of Windows NT Directory Services is for a user to have only one

tC
account and password, yet have access to all network resources, provided that the
user has been given the appropriate permissions. Although your network may
consist of multiple domains, the ability to establish trust relationships between
those domains is what enables users to have only one account yet still have
No
access to all network resources.
c

A user can log on to your network from any computer in a trusting domain by
using pass-through authentication. Pass-through authentication automatically for-
wards a user’s logon request to the appropriate domain for logon verification.
en
Do

Network Administration
After you have established trust relationships between your domains, you can
manage your network from any computer. The following table identifies some
se
r

common management tasks and the utilities you use to perform them.
fe

Utility Manages
ea

User Manager for Domains User and group accounts

Server Manager Domains and computers
Pl
Re

Both of these utilities enable you to manage all users and resources in all
domains on your network. To manage a different domain, simply select the appro-
priate domain within the utility.

30
You can create users and then use groups to help administer multiple users rather
than assigning resources and rights individually to each user, you can make
assignments to groups. Local groups contain users within a single domain. Global
groups are used to organize groups across multiple domains.

local group:
Groups of users in a single

ial
TASK 2C-1: NT domain.

INSTRUCTOR DEMONSTRATION—Creating users and

adding users to groups in NT global group:
Groups of users across
Objective: To use User Manager for Domains to create users and groups. multiple NT domains.

er
Setup: Observe as your instructor performs the steps in this task.

1. If necessary, log in as Administrator to the NTSRVR in the NETDOMAIN.

at
2. Choose Start→Programs→Administrative Tools (Common)→User Man-
ager for Domains.

y
3. Choose User→New User. In the New User dialog box, enter a username,
full name, and password. Uncheck User Must Change Password At Next

op
Logon. Click Add.
eM
4. Add another user (or more), and then close the New User dialog box.

tC
5. Next, let’s create a global group. Choose User→New Global Group. Name
the group GL-Students.

6. In the Not Members list box, select the student user names (Stu#) and
then click Add. Click OK.
No
c

7. Now, let’s create a local group. Choose User→New Local Group. Name
the group LCL-Group.
en

8. Click Add to display the Add Users And Groups dialog box. Select the user
Do

names created in this task, and then click Add. Notice that the Add
Names box contains the names in the format domain-name\user-name sepa-
rated by semicolons. Click OK twice.
se
r

Understanding Trust Relationships

In the Windows NT environment, you use trust relationships to establish secure
fe

links between domains. Trust relationships are configured through User Manager
ea

for Domains. When one domain trusts another, that domain’s users and global
groups are available from within User Manager for Domains and when configur-
ing share permissions. Once you have established trust relationships, you can give
Pl

users access to resources outside their domains. You can also manage all trusted
Re

domains from any computer.

Lesson 2: Servers and Clients 31

 Trusted and Trusting Domains
A trust relationship is a secure communications link between two domains so that
users from one domain may access resources in the other domain.
A trusted domain is a domain that is part of a trust relationship with another
domain. When a domain is trusted, the trusted domain’s users can be granted per-

ial
missions to other domains’ resources.
A trusting domain is a domain that makes its resources available to users from
other domains.
trusted domain:
A domain that is part of a
trust relationship with
Types of Trusts

er
another domain. When a
domain is trusted, the trusted
domain’s users can be The trust relationships between domains can be either one-way or two-way. In a
granted permissions to other one-way trust relationship, one domain trusts the other domain. In a two-way
domains’ resources.
trust relationship, both domains trust each other.

at
A one-way trust relationship is a trust relationship between two domains where
only one domain trusts the other domain. In a one-way trust relationship, the
trusting domain: trusted domain can access resources in both domains.

y
A domain that is part of a
trust relationship with A two-way trust relationship is a trust relationship between two domains where
both domains trust each other. In a two-way trust relationship, users from both

op
another domain. When a
domain is trusting, it makes
eM
domains can access resources in either domain.
its resources available to
users from other domains.

tC
Guidelines for Designing Trust Relationships
one-way trust When you design the trust relationships for your network, you should keep the
relationship: following guidelines in mind:
•
No
A trust relationship between You can only create trust relationships between Windows NT domains.
two domains where only one
c

domain trusts the other • You should establish the least number of trusts necessary for your environ-
domain. In a one-way trust ment to reduce the administrative overhead of managing those trusts.
relationship, the trusted
• The physical location of users does not matter since pass-through authentica-
en

domain can access resources

tion enables users to log on to the network across trusts.
Do

in both domains.
• Trusts are not inherited from domain to domain; they are non-transitive.
Consider the following example where the San Francisco domain trusts the
Los Angeles domain and the Los Angeles domain trusts the San Diego
two-way trust
domain. The San Francisco domain does not inherit a trust relationship with
relationship:
se

the San Diego domain from the Los Angles domain.

r

A trust relationship between

two domains where both
domains trust each other. In
fe

a two-way trust relationship,

ea

users from both domains can

access resources in either
domain.
Pl
Re

32
 ial
er
Figure 2-4: Trusts relationships are non-transitive.

at
Establishing Trust Relationships

y
When establishing a trust relationship, you must configure both domains from

op
eM
within User Manager for Domains to make the trust successful. You should con-
figure the trusted domain first, then configure the trusting domain. Although you
can initiate the trust relationship from either domain, you will receive an error
message if you start with the trusting domain because it will be unable to verify

tC
the trust. You will be able to use the trust immediately if you configure the
trusted domain first. However, if you configure the trusting domain first, it may
take up to 15 minutes for the trust relationship to be established.
No
c

TASK 2C-2:
en

Examining Windows NT Directory Services

Do

considerations in your network design

Objective: Read the following scenario, and then answer the questions
posed in the steps below. You’re planning to implement a net-
work consisting of 10 Windows NT servers, 5 NetWare
se
r

servers, a Microsoft SQL server with an important customer

database, and 2500 clients. You’d like to minimize administra-
fe

tive overhead.
ea

1. Define the following terms:

domain:
Pl
Re

trust relationships:

Lesson 2: Servers and Clients 33

 pass-through authentication:

2. With respect to management of user accounts, passwords, and security,

ial
what are some of the goals you might have?

er
3. How might you design the domain structure and, if necessary, the trust

at
relationships for this network?

y
op
eM
tC
Suggested time:
CHECK YOUR SKILLS 2-1
Under 5 minutes
Considering another scenario
No
c

Your network consists of two sites, one in San Francisco and the other in Los
Angeles. You have configured each site as a separate Windows NT domain. Your
user account has been created in the San Francisco domain.
en

How can you configure the network so that all users can log on from any
Do

computer at either site?

se
r

Windows NT Domain Models

fe

There are four different strategies you can use to design your domains and trust
ea

relationships. These strategies are referred to as domain models and include

single domain, single master domain, multiple master domain, and complete trust
domain.
Pl
Re

34
The Single Domain Model
In the single domain model, your network consists of only one Windows NT
domain. Thus, all user accounts and resources are contained within the same
domain. The single domain model supports the Windows NT Directory Services
goals of one account per user with access to all network resources and centralized
administration. Because the network consists of only one domain, you don’t need

ial
to configure any trust relationships.
When using this model, your network should have one Primary Domain Control-
ler (PDC) and at least one Backup Domain Controller (BDC). A single domain
can contain approximately 40,000 accounts in the directory services database.
Advantages of using this model include centralized administration of accounts

er
and resources and that it is the easiest model to implement because no trust rela-
tionships are required. Disadvantages of using this model include poor
performance if the domain is large and the need for powerful hardware for
domain controllers to maintain performance. Also, resources can’t be grouped as
they would be if they were in separate domains.

at
The Single Master Domain Model

y
The single master domain model consists of multiple domains. The master

op
eM
domain contains all user accounts for the network, and is therefore similar to a
single domain. The main difference between the two models is that the single
master domain model also contains resource domains. You must establish one-
way trust relationships between the master domain and the resource domains so

tC
that the user accounts can access any of the resources. After the trust relation-
ships are established, you can then give any user account access to resources in
any of the resource domains.
No
Each domain in your network should have a PDC and at least one BDC.
c
Although the model supports multiple domains, all user accounts are contained
within the master domain. Thus, you’re still limited to a maximum of 40,000
users in the directory services database.
en

Advantages of this model include centralized administration of user accounts and

Do

decentralized administration of network resources. Resources can be grouped into

domains either by location or by department.
Disadvantages of this model include poor performance if the domains are large,
each resource must have its own local groups for granting permissions to
resources, and that only the administrators of the master domain can create users
se
r

and global groups.

fe
ea

The Multiple Master Domain Model

If your network requires more than 40,000 user accounts, then one design strat-
egy you can use is the multiple master domain model. In this model, you
Pl
Re

configure two or more master domains and multiple resource domains. Together
the master domains contain all of the user accounts for the entire network. The
resource domains contain the network resources and are usually defined based on
location or department.

Lesson 2: Servers and Clients 35

 To maintain the goal of a single user account for each user for the entire network,
the master and resource domains must be able to exchange user information
through trust relationships. Each master domain must have a two-way trust rela-
tionship with all other master domains and each resource domain must have a
one-way trust relationship with all master domains; the resource domain must
trust the master domains.

ial
Advantages of this model include:
• Accommodation of more than 40,000 users.
• Centralized administration of user accounts.
• Decentralized administration of network resources.
• Resources can be grouped into domains either by location or department.

er
Disadvantages of this model include:
• It is more complicated to design and manage because of the large number of
trust relationships.
• User accounts are split over multiple domains.

at
• It must create global groups in each of the master domains.

y
The Complete Trust Domain Model

op
eM
In the complete trust domain model, all domains on your network are connected
via two-way trust relationships. This model is similar to the multiple master
domain model, in that it supports networks with more than 40,000 users. How-

tC
ever, the administration of a complete trust network is completely decentralized.
Advantages of this model include:
• It can accommodate more than 40,000 users.
No
• It accommodates companies that don’t have a central network management
c

staff because the administration of network resources and user accounts is

decentralized.
• It can group resources into domains either by location or by department.
en

•
Do

Each domain has complete control over its user accounts and resources.
Disadvantages of this model include:
• It is more complicated to design and manage because of the large number of
trust relationships.
•
se

It doesn’t accommodate companies with a central network management staff

r

because the administration is decentralized.

•
fe

It may have security problems if domain administrators put inappropriate

ea

global groups into their local groups.

Pl
Re

36
TASK 2C-3:
Discussing domain models
1. What are some benefits to implementing the single domain model?

ial
2. How does administering a single master domain model differ from
administering a single domain model?

er
at
3. Why would you choose to implement the multiple master domain model
instead of the single master domain model?

y
op
4.
eM
How should you determine where to create your users’ accounts in the
multiple master domain model?

tC
No
c

5. When using the multiple master domain model, where should you create
the necessary global groups?
en
Do

6. Why would you choose to implement the complete trust domain model?
se
r
fe
ea

7. What is a risk of implementing the complete trust domain model?

Pl
Re

Lesson 2: Servers and Clients 37

 Novell Directory Services
Novell Directory Services (NDS) is a major feature of NetWare, starting with
NetWare 4.0 and continuing through the current versions of 4.x and 5.x. Previous
versions of NetWare were bindery-based (a flat-file database structure).
NDS (Novell Directory
Services): NDS maintains information about all network resources (such as servers, vol-

ial
Formerly known as NetWare umes, users, groups, and printers) in a hierarchical, tree-structured database that
Directory Services. An resembles an organizational chart. With it, system administrators can organize
administrative database on network resources according to the way a given work group is structured—an
the network that defines efficient method for giving users access to the network according to the role they
network entities, such as play in the organization. When a user requests a resource, NDS locates the
users, groups, servers, and
so on. NDS was introduced
resource, verifies that the client is authorized to access the resource, and makes

er
with NetWare 4.0, and the resource available to the client.
controls access to network
information, resources, and
services. It allows users to
access all network resources Examining the NDS Database and its Components
with a single login

at
procedure. NDS enables users to access network information and resources. Every resource
on the network has an entry in the NDS global database. This database, called the
Novell Directory, represents network resources as database objects that can be

y
organized independently of physical location. All NetWare 4.x and 5.0 servers
connected to a single network have access to, and use, the same Novell Directory.

op
eM
Therefore, they have information on all available resources on the network. Thus,
a single point of access exists for both using and managing network resources.
When you want to access a resource, it is not necessary for you to know the
name of the network server providing that resource. To gain access or receive

tC
The NDS tree is also called information about the resource, you simply identify it by name, provided that you
the Directory tree. You might have the proper rights to it.
encounter the term Directory
tree in NetWare-related
documents.
No
The NDS Tree
c

Because NDS is critical to network management and security, you must carefully
plan the NDS structure. The Directory objects are organized in a hierarchical
structure called the NDS tree ( or Directory tree ). The NDS tree is similar to the
en
Do

tree structure found in DOS file systems.

se
r
fe
ea
Pl
Re

38
Like the DOS root directory, the [Root] object exists at the top of the tree. Con-
tainer objects, analogous to DOS directories, can exist within either the [Root]
object or each other. Leaf objects, the structural equivalent of DOS files, can be
placed within Container objects. [Root] object:
Represents the top of the
The NDS tree can contain an unlimited number of container levels, so you can NDS tree and is created
set up a logical structure that reflects your organization’s network environment. during the installation of the

ial
However, you should try to keep the number of levels in your tree to a minimum; NetWare network operating
otherwise, they might become unwieldy to administer. NetWare documentation system.
indicates that you should not have more than six levels in your NDS tree.
The Novell Client software enables you to browse the NDS tree using Windows
Explorer and Network Neighborhood. It is the only client software for Windows Container object:
95 and Windows NT Workstation that provides full support of NDS. An object that can contain

er
other objects and is used to
Note: The NDS tree, unlike a DOS directory tree, has restrictions on where specific organize the Directory tree.
Container objects can be placed.

at
Leaf object:
An object that can’t contain
other objects, such as a User
object.

y
op
eM
tC
No
c
en
Do
se
r

Figure 2-5: NDS Directory structure.

fe

You can set up an NDS tree in different configurations. Some examples include:
ea

• Shared resources. If personnel from different departments need to access a

particular printer or server, you can arrange the NDS tree to provide access.
Consider carefully how you
• Job levels. Vice-presidents, division managers, and department managers can design the NDS tree. The
Pl

be grouped together along with their respective resources. stability of your company
Re

organizational chart and floor

• Plant location. If the company has plants in several locations (on streets plan should also be taken
such as Corporate Boulevard, Winding Way, and Elm Street), you could into consideration when
organize the NDS tree by using Container objects named CorpBlvd, evaluating examples such as
WindingWay, and Elm. those in this list.

Lesson 2: Servers and Clients 39

 • Floor. If departments located on the same floor in the company share many
of the same resources, you might organize the NDS tree by floor or by part
of the building.

Tree Design

ial
You should strive for a tree design that is easy to administer and that provides
users with access to the resources they need. NDS trees can be based on geo-
graphic location, organizational structure, administrative responsibilities, or any
combination of these.
This flexibility in tree design also brings with it added complexity, so careful

er
planning before you implement your NDS tree can be one of the most important
steps in implementing a NetWare network.
Other factors that can affect a tree design include workflow, resource allocation,
network traffic, fault tolerance, and the flow of information through an
organization.

at
TASK 2C-4:

y
op
eM
Discussing tree-design considerations
1. What are some other ways the NDS tree in Figure 2-5 might have been

tC
structured? No
2. What is the main function of the NDS tree?
c

3. What two things should you ensure that a tree design facilitates?
en
Do

4. How can NDS trees be organized?

se
r

5. List some other things that might affect a tree design.

fe
ea
Pl
Re

Objects
In NDS, an object represents a network resource and provides a way to identify
and manage the network resource it represents. You use objects to organize, find,
access, and manage network resources, such as users, servers, and printers.

40
Each NDS object has a name, a list of properties (a collection of information that
describes the object), and specific values for each of its properties.

ial
er
Figure 2-6: An object consists of properties and their values.
In Figure 2-6, the User object BBerkley represents that user on the network and
holds information about him. By configuring the User object’s properties (such as

at
whether it requires a password), a system administrator can control how the user
accesses the network. The object can also store reference information, such as the
user’s phone number.

y
op
eM
Properties and Values
Properties are the categories that can store information for an object. Values are

tC
the information stored in the property fields. NDS stores information about net-
work resources much like information is stored in a database.
No
c
en
Do
se
r
fe
ea
Pl
Re

Figure 2-7: Databases.

Lesson 2: Servers and Clients 41

 Some properties store information that is essential to the network and to the sys-
tem administrator. For instance, a user’s login name is critical in allowing that
user network access. Other properties might contain non-essential information,
NDS defines the properties such as a user’s title, telephone number, or fax number.
for each object. Developers
can use third-party APIs to Some properties have mandatory fields for which you must enter a value. For
develop additional properties example, when you create a User object, Last Name is the only required property.

ial
for an object.
Other properties can hold multiple values. For example, the Telephone Number
property can contain several values (that is, several phone numbers where the
user can be reached).

er
at
y
op
eM
tC
Figure 2-8: An NDS tree showing the [Root] object, Container objects, and Leaf objects.
No
c

Object Types
As its name implies, the NDS tree is a hierarchical tree structure in which NDS
en

objects are stored. NDS objects are divided into three classes: the [Root] object,
Do

Container objects, and Leaf objects.

The [Root] Object

se

During the installation of the first NetWare 4.x or 5.x server on the network, the
r

installation program creates the [Root] object. The [Root] object can contain only
the Novell Directory and Country, Organization, and Alias objects. You must use
fe

square brackets when referring to the [Root] object.

ea

Container Objects
Pl
Re

Container objects enable you to organize the Novell directory logically, much as
folders and drawers organize the documents in a filing cabinet. They can repre-
sent countries, companies, divisional units, departments, responsibility centers,
workgroups, or shared resources.

42
Container objects can hold other Container objects or Leaf objects. Each class of
Container object differs from the others by a set of hierarchical rules that define
what it can contain and where it can exist. The classes also differ in what infor-
mation their properties store.
The following table describes Container objects, including where they can exist
and what objects they can hold.

ial
Icon Name Description Can exist in Can hold
Country Represents the country in [Root] Organization
which the network is located. objects.

Organization Organizes other objects in the [Root] object or Organizational

er
NDS tree. For example, it might Country objects. Unit objects and
represent companies or all Leaf objects.
departments.
Organizational Organizes other objects in the Organization or Organizational
Unit NDS tree. For example, it might Organizational Unit objects and

at
represent departments, Unit objects. all Leaf objects.
business units, project team, or
divisions.

y
op
eM
tC
No
c
en
Do

Figure 2-9: Container objects in an NDS tree.

se
r

See if you can figure out the

container types in this figure.
fe
ea
Pl
Re

Lesson 2: Servers and Clients 43

 Leaf Objects
Leaf objects represent the network’s resources. They cannot contain other objects.
The following table describes the Leaf objects created by default during server
installation:
You can use Novell online
documentation to search for
Icon Leaf object Description

ial
the phrase Leaf objects. The
results of this search will list NetWare Server A server running any version of NetWare starting at 4.x or
Object Classes And higher.
Properties and this section
includes Leaf Object Classes. User Represents a person who can use your network.
This link displays a list of
Leaf objects and the icons Volume Represents a physical volume.

er
associated with them.

Default NDS Objects

at
The following NDS objects are automatically created during the installation of
NetWare 4.x and 5.x:

y
• A NetWare Server object for the server on which you installed NetWare.

op
•
eM
The SYS Volume object, which contains the NetWare operating-system files.
• Any other Volume objects that you created during installation.
• The User object ADMIN , which initially has complete access to all objects

tC
in the tree. An administrator uses this account to perform the initial network
setup.
No
NDS Utilities
c

NetWare provides several utilities for working with the NDS tree. They include:
en

Bindery-based NetWare
Do

operating systems use the

SYSCON utility to manage
users and groups,
PCONSOLE to manage print
services, and FILER to
manage the file system.
se
r
fe
ea
Pl
Re

44
• NetWare Administrator (NWADMN32.EXE), which provides a Windows-
based graphical environment in which you can create NDS objects and
manage the Novell directory. You can also use it to perform printing and
file-management functions. This file is located in the SYS:PUBLIC\WIN32
directory.

ial
er
at
y
op
eM
Figure 2-10: Viewing the NDS tree by using NetWare Administrator.

tC
• NDS Manager, a utility for performing advanced database-management tasks,
such as partitioning and replicating the Novell directory.
• ConsoleOne , a Java application for managing network resources is included
with NetWare 5. ConsoleOne is platform-independent, which means that you
No
can run it from any client or from the server itself.
c

• In NetWare 3.x you use the SYSCON utility used instead of NetWare Admin-
istrator or ConsoleOne to administer the bindery objects (users and groups).
en
Do

Partitions and Replicas

Part of administering the NDS database is making the database secure. This is
done by dividing the NDS database into more manageable pieces, called parti-
tions, and providing copies of those pieces, called replicas, on servers throughout
se
r

the network. Proper placement of these partitions and replicas will help optimize
NDS management, workgroup management, and user access; provide a backup of
fe

NDS, and reduce unnecessary traffic; and eliminate any single points of failure.
ea

Proper partitioning and replication is part of completing your NDS design

structure.
Pl
Re

Partitions
Partitions are logical sections of the NDS database. The NDS database can
become very large. By sectioning the NDS database and placing these sections on
different servers, you can manage the database more efficiently.

Lesson 2: Servers and Clients 45

 Replicas
A replica is a copy of all the data (objects and all their information and values)
stored on a server within the boundary of a partition. Replicas are hidden files
stored on the SYS volume on servers throughout the network and are managed
by administrative users. You should have at least three copies of each partition.
(If you have only one or two servers, you cannot follow this guideline.) Each

ial
partition should have no more than five replicas. Over-replication takes up too
much bandwidth, due to replica synchronization.
Replicas serve two primary purposes:
• Provide fault tolerance; and
• Enhance network performance.

er
Replica Types
There are four types of replicas:

at
• Master replica . The original and only complete copy of a partition. Users
can authenticate using a master replica.

y
• Read/Write replicas. A replica designed for users to view and search for
object information. Users can authenticate using a read/write replica.

op
•
eM
Read-Only replicas. A replica designed for users to view and search for
object information.
• Subordinate references. A pointer created by NDS to point to other replica

tC
types where parent partitions exist without their child partitions.

TASK 2C-5:
No
c

Examining objects in the NDS tree

en

1. Log in to the 411SRVR as Admin. This is the classroom NetWare 4.11

Do

server. Admin is the default administrative user on a NetWare 4.11 network.

2. Start NetWare Administrator by choosing Start→Run, then browsing to

the 411SRVR, then to the PUBLIC\WIN95 folder, and running
NWADMN95.EXE.
se
r

3. Identify the container objects in the browser window. Double-click on

the container objects to expand them and display the Leaf objects below
fe

the containers.
ea

4. Identify the Leaf objects in the browser window. Double-click on a Leaf

object. The properties of the object are displayed in a separate window.
Click Cancel to return to the Browser window.
Pl
Re

5. Explore the properties and values for the Admin object, making sure not
to make any changes to the object.

46
6. Let’s create a user. Select the NetPlus container. Choose Object→Create.
In the Class Of New Object list box, select User, and then click OK. In the
Login name text box, enter your name; then in Last Name, enter your last
name. Click Create. Verify that the user is in the NetPlus container.

7. Close NetWare Administrator and log out.

ial
Summary
In this lesson, you’ve examined the three most popular network operating

er
systems, identified features of network clients, and examined the directory
services used by NT and NetWare networks.

at
LESSON 2 REVIEW

y
2A What are three popular network operating systems?

op
eM
2B What role does the client have in a client-server network?

2C What goals do directory services try to attain?

tC
No
c

What is the Windows NT 4 directory service called and how is it imple-

mented?
en
Do

What is the Novell NetWare directory service called and how is it imple-
mented?
se
r
fe
ea
Pl
Re

Lesson 2: Servers and Clients 47

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Introduction to the OSI LESSON
Model
3

ial
Data Files:
Network+ Certification
Overview CD-ROM

er
In preparation for taking the Network+ exam, you need to have an under- Lesson Time:
standing of how various devices interrelate. The OSI model breaks these 1 hour
communications into layers. This lesson explains those layers and describes
their functionality.

at
Objectives

y
To grasp the details of the OSI model and its relation to most networking compo-

op
nents, you will:
eM
3A Identify the OSI layers.

tC
You’ll examine the importance of the individual layers of the OSI model.
The Application, Presentation, Session, Transport, Network, Data Link,
and Physical layers are the seven layers in the model.

3B Describe relationships between families of protocols.

No
c

You’ll identify the types of protocols that are found in Network and
Transport layers, and in the top three layers (Application, Presentation,
and Session).
en
Do
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 49

 TOPIC 3A
Introduction to the OSI Model
The OSI (Open Systems Interconnect) reference model was developed in the

ial
early 1970s by the International Standards Organization (ISO). The OSI model
was designed to provide a structured and consistent approach for describing,
understanding, and implementing networks. Seven individual layers make up the
structure of the OSI model. Those seven layers in order from highest to lowest
are Application, Presentation, Session, Transport, Network, Data Link, and
Physical. Before the implementation of the globally recognized OSI model, the

er
only solution for an enterprise wide network was to use a single vendor. The OSI
model provides us with a set of general design guidelines for data-
communications systems and also gives us a standard way to describe how
various portions (layers) of data-communication systems interact. Vendors now
use the OSI specifications as a guideline to their network products.

at
y
op
eM
tC
No
c
en

Figure 3-1: Data flow on a simple network architecture.

Do

The various layers represent the data flow pattern that will take place on even the
most basic computer network. Communication takes place between the various
layers in both directions as data moves up or down through the architecture (see
Figure 3-1). Data starts at the Application layer and is sent down through the lay-
ers to the Physical layer. The receiving computer picks up the data at the Physical
se
r

layer and sends it up to the Application layer; this process is repeated over and
over.
fe
ea
Pl
Re

50
Layered Communication
The various layers in a network architecture represent the different stages and
types of processing that occur when information moves through a computer
network. Consider the analogy shown in the following figure, in which Albert
sends a letter to a colleague in Germany. Certain activities occur in each layer to
ensure that Albert’s message is delivered accurately, correctly, and quickly. Com-

ial
munication occurs between the layers and the message moves down or up
through the architecture.
Be sure to notice that when
the sender addressed the
letter to Germany he did
request a return receipt. After
the letter is delivered, a

er
receipt will be returned to the
sender to acknowledge
successful delivery. This can
also occur in network
environments.

at
y
op
eM
tC
No
c
en
Do

Figure 3-2: In network architectures, messages pass through several layers, as in this
analogy.

The Seven Layers of the OSI Model

se
r

Each of the seven layers in the OSI model have a specific set of responsibilities
fe

that allows for reliable network communication. These layers and their general
ea

purpose are described in the following figure.

Pl
Re

Lesson 3: Introduction to the OSI Model 51

 ial
er
at
y
op
eM
Figure 3-3: Layers of the OSI model.
tC
The above layers are numbered from 1 for the Physical layer to 7 for the Appli-
No
cation layer. The responsibilities of each of the layers are briefly summarized
c

below.
Remembering the seven layers can be the toughest thing to do in order to learn
the model. Perhaps the best method of learning these layers is to come up with a
en

memorable mnemonic device. Probably the two best examples of mnemonic

Do

A mnemonic device for devices are All Pizzas Seem To Need Double Pepperoni , and All People Seem To
remembering the layers in Need Data Processing . It may be more beneficial to come up with your own
their proper order is All
mnemonic device so you can remember it come test time. Make it something
Pizzas Seem To Need Double
Pepperoni. funny or catchy and you’re bound to have a great memorization tool.
se
r

Responsibilities of Network Layers

fe
ea

As you learn the responsibilities of each layer, the most important concept to
keep in mind is that each of the layers performs distinct functions, working with
the layers above and below it to enable network communications. Any one layer
on its own is largely useless. And, no layer infringes on the responsibilities of
Pl
Re

any other layer—they don’t do each other’s jobs.

It’s also important to remember that the OSI model layers are only architectural
models and don’t actually perform tasks. Rather, protocol schemes perform the
tasks that correspond to each layer.

52
Same-stack Interaction
The hierarchical layering of protocols on a computer that forms the OSI model is
known as a stack. Typically, each computer has one. Protocols for each layer in a
computer’s stack communicate with their counterparts in another computer’s
stack. To communicate, however, messages must flow down through the lower
layers of the sending stack. A given layer in a stack sends commands to layers

ial
below it and services commands from layers above it. The layers use
preconfigured addresses, known as service access points, to communicate infor-
mation to each other.

Inter-stack Interaction

er
When layers in two different stacks communicate, a layer in one stack communi-
cates with its counterpart in the other stack. It does so by including its own
information with the data information it sends. This information is called a
header, which includes information and any requests by the layer. The layer’s

at
counterpart in the other stack receives, reads, and then processes the header, as
shown in Figure 3-4.

y
op
eM
tC
No
c
en
Do
se
r

Figure 3-4: Sending layers communicate with their counterparts on the receiving stack by
fe

using headers.
ea

When a layer sends its header and the data information to the layer below it, the
lower layer treats all the information as one data packet and attaches its own
header to it. When the data packet reaches its destination in the other stack, the
Pl

original layer’s counterpart recognizes which header it should acknowledge

Re

because lower-layer headers have been stripped off the packet by those layers.
Unlike the downward flow of the data packet in the sending stack, the flow of the
packet in the receiving stack is upward. The data flows up the layers with each
header being stripped off in a first-in, last-out fashion.
In summary:

Lesson 3: Introduction to the OSI Model 53

 • As information moves down the layers, each layer might add its own header
to tell the corresponding layer at the destination how to deal with the data.
By the time the information reaches the Physical layer, five headers could
have been added to it (one for each layer except for Application and
Physical).
• As information moves up the layers at the destination, each layer removes

ial
the header that was added by its peer (the same layer) on the transmitting
end and processes the information contained within that header. When the
information reaches the Application layer, it should be the same as the origi-
nal message. The primary purpose of headers is to enable transmitting layers
to communicate with their peer layers on the receiving end.

er
Service Data Units
The data transferred between layers is known as a service data unit, preceded by
a given layer name. For example, data information given to layer 3 from layer 4

at
is called a Network layer service data unit.
service data unit:
The data transferred between Service data units are made up of data bits, which are referred to by different
names, depending on which layer you are talking about. The following list con-

y
layers of the OSI model.
tains widely used names and the layer to which they pertain. These names,

op
however, are not set standards and are often misused.
eM
• Bits: Physical layer
• Frames: Data Link layer
•

tC
Datagrams and segments: Transport layer
• Messages: Application layer

Note: Packet, among other terms, is also used at many model layers.
No
c

The Application Layer

en

The Application layer includes network software that directly serves the user, pro-
Do

viding such things as the user interface and application features. This layer is
often times mistaken for an actual software package, such as a spreadsheet
program. Do not do this. The Application layer is usually made available by

Application layer:
se
r

Level 7, the highest layer in

the OSI model. It provides a
user interface.
fe
ea
Pl
Re

54
using an Application Programmer Interface (API), or hooks, which are made
available by the networking vendor. This layer doesn’t include application soft-
ware such as Microsoft Word or Excel. However, it does enable commercial
software to use network services such as printing and message handling. Application
Programming Interface
(API):
Application Programming

ial
Interface. An extension to a
programming language that
defines an interface to an
operating system. If an
application is written for a
particular API, it need only
be recompiled to run on any

er
other operating system that
provides that API.

at
y
op
eM
tC
Figure 3-5: The Application layer.

The Presentation Layer

No
c
Software in the Presentation layer attempts to resolve differences in the way data
is represented (character encoding, bit order, and byte order issues) among net-
work computers, for not all computers represent data in the same way. This layer
translates data to ensure that it is presented properly for the end user (in the
en

Application layer). The Presentation layer also handles related issues such as data
Do

encryption and compression, and how data is structured, as in a database. Presentation layer:
Layer 6 in the OSI model. It
provides code and syntax.
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 55

 ial
er
at
y
op
eM
Figure 3-6: The Presentation layer.

The Session Layer

tC
Session layer software establishes and maintains the connection between applica-
tions that communicate with each other through the network. To do this, the
Session layer must employ services provided by the Transport layer. The Session
No
layer comes into play primarily at the beginning and end of a transmission. At the
c

beginning of the transmission, it makes known its intent to transmit. At the end
Session layer: of the transmission, the Session layer determines if the transmission was
Layer 5 in the OSI model. It successful. This layer also manages errors that occur in the upper layers, such as
en

provides a logical a shortage of memory or disk space necessary to complete an operation, or

connection.
Do

printer errors. Network computers can interact on a hierarchical, client/server, or

peer-to-peer basis. Typically, the type of relationship that exists between comput-
ers is established in the Session layer.
se
r
fe
ea
Pl
Re

56
 ial
er
at
y
op
Figure 3-7: The Session layer.
eM
The Transport Layer
The Transport layer provides the upper layers (Application, Presentation, and
tC
Session) with a communication channel to the network. To do this, the Transport
layer uses the services provided to it by the lower layers (Network, Data Link,
No
and Physical). The Transport layer is aware of the packet size that lower layers
c

can handle. When sending a message, the Transport layer determines whether the
message is too big or too small to transmit. If it is too big, the message is broken Transport layer:
down into smaller pieces. If the message is too small, the Transport layer looks to Layer 4 in the OSI model.
en

see if there are other small pieces that are going to the same place, and combines This layer is the heart of the
Do

them when sending. When receiving a message, the Transport layer collects and model; it provides reliability.
reassembles any packets, organizing the segments for delivery and ensuring the
reliability of data delivery by detecting and attempting to correct problems that
occurred (such as missing or improperly sequenced packets of data). If the Trans-
port layer cannot correct data-transmission problems, then it sends a message to
se

the upper layers that a problem has occurred.

r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 57

 ial
er
at
y
op
eM
Figure 3-8: The Transport layer.

The Network Layer

tC
Networks, especially large internetworks, might provide more than one path for
moving information from one point in the network to another. The Network lay-
er’s main purpose is to decide which physical path the information should follow
No
from its source to its destination. The path might span the distance between two
c

computers in the same office, or it might span the globe.

Network layer:
Layer 3 in the OSI model.
en

This layer deals with routing

Do

and addressing.
se
r
fe
ea
Pl
Re

58
 ial
er
at
y
Figure 3-9: The Network layer.

op
eM
The Data Link Layer

tC
Most networks require that only one device transmit on the communication chan-
nel at a time. The Data Link layer provides a system through which network
devices can share the communication channel. This function is called media-
access control (MAC) . The Data Link layer also uniquely identifies each
No
computer on the network, usually by a pre-encoded address that is part of the
c

unit’s network interface card. This pre-encoded address is typically called a MAC Data Link layer:
address and is said to be ″burned in″ by the manufacturer, as they are the ones Layer 2 in the OSI model. It
who do the pre-encoding. shares the media and
en

controls the flow.

The Data Link layer prepares data for transmission over the communication
Do

channel. On the transmitting end, the Data Link layer arranges signals into logical
units of information called frames , which it passes to the Physical layer for
transmission through the network. To ensure that frames are an appropriate length
for the Physical layer, the Data Link layer can fragment data that comes from the
layers above it. On the receiving end, the Data Link layer reassembles the frag-
se
r

mented data and attempts to detect and correct transmission errors that occurred
in the Physical layer. The Data Link layer also establishes and maintains the con-
fe

nection between one device to another while transmission is taking place. This is
ea

called logical link control (LLC) .

To represent the two major functions of the Data Link layer, it is most often con-
sidered to be divided into two sub-layers: the MAC layer and the LLC layer.
Pl
Re

Lesson 3: Introduction to the OSI Model 59

 ial
er
at
y
op
eM
Figure 3-10: The Data Link layer.

The Physical Layer

tC
The Physical layer further addresses the hardware used to transmit data over the
network. This includes the voltage used, the timing of data transmission signals,
No
the requirements for establishing the initial communication connection, acceptable
c

connectors, and interfaces to data-transmission media. The Physical layer provides

the electro-mechanical interface through which data moves among devices on the
Physical layer: network. It specifies the type of data-transmission medium used to link network
en

Layer 1 in the OSI model. It devices. It also specifies how the medium interfaces with network devices, and
Do

is the physical and electrical how data is encoded into electronic signals for transmission through the medium.
interface. Finally, the Physical layer specifies what constitutes an acceptable signal.
se
r
fe
ea
Pl
Re

60
 ial
er
at
y
Figure 3-11: The Physical layer.

op
eM
Making Connections
The Physical and Data Link layers provide local connections, which connect a

tC
node to other devices on the same network segment.
The Network and Transport layers enhance the services provided by the two
lower layers to provide remote connections or internetwork connections, which
link nodes. This type of connection is sometimes called an end-to-end connection.
No
c

The Session layer enhances the services provided by the lower four layers to pro-
vide logical connections, which establish a link between communicating processes
on separate nodes, as shown in Figure 3-12.
en
Do
se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 61

 ial
er
at
y
op
eM
Figure 3-12: Making connections.

tC
A Network in Action
As mentioned previously, the protocols in the OSI model layers communicate
with each other to complete tasks. It is similar to sending a letter to a friend. The
No
communications involved in that transaction extend well beyond you and your
c

friend. For example, when you place the stamped and addressed letter in your
mailbox, you communicate to the postal carrier that you want the letter taken to
the post office for sorting and eventual delivery to your friend. The postal carri-
er’s placement of the letter in a bin at the post office communicates to the postal
en

clerk that the letter needs to be sorted. Several other transactions need to occur
Do

before your letter reaches its destination.

Figure 3-13 depicts a simple exchange in which a network node requests a file
from a file server. Network components are shown in relation to the OSI refer-
ence model. As shown here, it is not unusual for some components to span more
se

than one layer of the OSI model to accomplish the communication task at hand.
r

A. An application running on the workstation requests a file that is stored on

fe

the network.
ea

B. Redirection software on the workstation determines that the request is for

network services. It directs the request to the communication protocol for
transmission on the network.
Pl
Re

C. The communication protocol determines the best route through the network
and passes the request to the LAN driver, along with instructions that the
message should be delivered to the file server.
D. The LAN driver copies the request into frames (bundles that can be moved
across the network) and sends the frames to the network adapter for
transmission.

62
E. The workstation’s network interface board transmits the frames across the
network.
F. The file server’s network interface board receives the frames and passes
them to its LAN driver.
G. The file server’s LAN driver takes the message out of the frames and sends
a confirmation to the workstation that it has received the frames.

ial
H. The communication protocol verifies that the message was received intact,
and it passes the message to the network operating system.
I. The network operating system receives the request and acts on it. The net-
work operating system will use the layers of the network to transfer the
requested file from a network hard disk to the workstation.

er
at
y
op
eM
tC
No
c
en

Figure 3-13: Components in a simple exchange between a workstation and a file server,
Do

shown in relationship to the layers of the OSI model.

se
r
fe
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 63

 TASK 3A-1:
Viewing a model of a network in action
Objective: To view the presentation (on the CD included with your
manual) that shows the layers of the OSI model and how the

ial
OSI model layers interact.
Setup: Access to the Network+ Certification CD-ROM should have
been configured on your workstation prior to class.

1. Following your instructor’s directions, access the Network+ Certification

er
CD-ROM. By default , choose Start→Programs→Network+ Interactive
Learning→Network+ Interactive Learning CD.

2. From the Main Menu, click on the Digital Video link.

3. Click the Network Cables And Devices link, and then the OSI Model

at
link.

4. Click on the Overview link to start the presentation.

y
5. View and listen to the digital video presentation.

op
eM
6. Return to the Main Menu and Exit the Network+ Certification
CD-ROM, back to your desktop.

The Benefits of the OSI Model

tC
By separating the tasks necessary for network access into a number of distinct
No
layers, the OSI Model enables the upper layers to work independently of the
c

lower layers. For users, this translates (at least theoretically) to network
transparency. Due to layering, your application should work equally well over an
Ethernet network as over a Token Ring or ARCNet network. Other than speed
en

differences, the networks should work identically because the OSI Model enables
Do

the components (the implementations of a layer or layers) to work independently.

Again, in theory, on a workstation to switch from an Ethernet to a Token Ring
network, you should have to switch only the network interface card (which takes
care of the Physical and some of the Data Link functions) and some of the driv-
ers (to take care of the remainder of the Data Link layer). The rest of the layers,
se

and the drivers that implement them, should be able to remain in place without
r

modification.
fe

The ISO’s publications describe the specifics of each layer in great detail. Even
ea

so, variations exist between network implementations that purportedly follow the
OSI Model. Only your testing can ensure that OSI-compliant systems are truly
compatible for your environment. However, on their own, the ISO’s detailed
specifications go a long way toward ensuring compatibility.
Pl
Re

64
TASK 3A-2:
Identifying layers of the OSI model
1. List the seven layers of the OSI model and the function of each layer.

ial
er
at
2. Create a mnemonic device to help you remember the layers (in order) of

y
the OSI model.

op
eM
tC
APPLY YOUR KNOWLEDGE 3-1 Suggested time:
10 minutes
Developing an analogy for the OSI model
No
Using the analogy in Figure 3-2 as a model, create your own analogy for the OSI
c

model and identify the functions of each layer. The layers in your analogy should
correspond to OSI’s layers. Don’t use data communications for your analogy.
Be sure to include what would happen at each layer of the OSI model:
en
Do

Application layer
Presentation layer
Session layer
Transport layer
se
r

Network layer
Data link layer
fe

Physical layer
ea
Pl
Re

Lesson 3: Introduction to the OSI Model 65

 TOPIC 3B
Families of Protocols
Network communication is provided through protocols —rules by which network

ial
operations are conducted. Network protocols are like species of animals: there are
many of them, and they vary endlessly. Like animal species, protocol ″species″
can be classified into general families that share common features. Some proto-
cols are hybrids; they perform more than one function. Others perform only one
particular function. Computer technology professionals use the OSI model to clas-
sify protocols by function. The following sections introduce families of network

er
Make sure that you protocols and their respective OSI model layers.
understand that OSI model
layers don’t perform network
functions; rather, the
protocols associated with
each layer perform them.
Network and Transport layers

at
The Network and Transport layers provide the following protocol families:
• Reliability protocols provide a way to ensure that reliable data transfer
occurs. For example, a header or trailer might contain a checksum or a

y
request that received data be acknowledged by sending an acknowledgment
message back to the sender.

op
•
eM
Connection protocols are used for establishing and maintaining a
connectionless or connection-oriented service for upper layers. In a
connection-oriented service, the sending and receiving stations maintain con-

tC
stant communication to mediate the transfer of data. Sequencing, flow
control, and reliability are monitored by both ends. This is like a telephone
call in which two parties interact constantly. Connectionless service is more
like the postal service. The message is transferred with little, if any, commu-
No
nication between the two parties, other than the message itself. The message
c

is packaged, delivered, and sent.

• Routing protocols provide a way to ensure that data is transferred to the cor-
rect destination. In an unswitched network, routing is virtually unnecessary,
en

because nodes are directly connected. In a switched network, however, the

Do

routing function determines which path a data unit will take to reach its
destination. This function is particularly important and complex in a packet-
switched network, because there can be many possible paths to a destination
and many intermediary nodes (called routers ) along the way. Routing proto-
cols determine the strategies used to transmit data through the network.
se
r
fe

Application, Presentation, and Session Layers

ea

The Application, Presentation, and Session layers provide the following protocol
families:
• Terminal-emulation protocols enable computers to behave like a standard
Pl
Re

terminal so that they can access a host. This typically involves translation of
keyboard codes and video-display codes.
• File-access protocols enable nodes to use network files. For example, differ-
ent network clients might use different file- and path-naming conventions.
File-access protocols provide a common means to access network files.

66
• File-transfer protocols copy files between network storage and other storage,
such as a computer’s local disk drives.
• Email protocols provide for electronic-mail delivery and handling of related
messages.
• Remote-action protocols determine whether processes should be performed
by a client node or by a server. These protocols are required for setting up a

ial
client-server relationship.
• Network-management protocols provide tools for setting up and maintaining
the network. As networks interconnect into internetworks and become more
complex, more sophisticated network-management tools are necessary.
• Task-to-task protocols enable software processes to communicate over the

er
network.
• Multiple-session protocols enable multiple network links to be established.
• Codeset-and-data-structure protocols define how data is represented. These
protocols translate data for nodes that use different coding schemes.

at
TASK 3B-1:

y
Identifying families of protocols

op
eM
1. Identify the family of protocols that determines whether a client or
server completes a process.

2. Identify the family of protocols that ensures that data is transferred

without errors.
tC
No
c

3. Identify the family of protocols that enables a computer to access a host

node by behaving like a terminal.
en
Do

4. Identify the family of protocols that provides tools for setting up and
maintaining the network.
se
r

5. Define the function of the routing family of protocols.

fe
ea

6. Define the function of the file access family of protocols.

Pl
Re

Lesson 3: Introduction to the OSI Model 67

 7. Define the function of the codeset and data structure family of protocols.

8. Define the function of the connection family of protocols.

ial
Summary

er
In this lesson, you’ve identified the layers of the OSI reference model and
learned the functions of each of those layers. You also learned about the
kinds of protocols that work at each of the OSI layers.

at
LESSON 3 REVIEW

y
op
eM
3A Describe the purpose of network architecture and the OSI model.

tC
List the OSI layers and describe how they communicate.
No
c

3B Describe what is meant by the families of protocols.

en
Do
se
r
fe
ea
Pl
Re

68
The Physical Media LESSON
4

ial
Overview Data Files:
This lesson will concentrate on the Physical layer of the OSI model. The Network+ Certification
Physical layer describes the physical and electrical characteristics of the net- CD-ROM

er
work medium. It deals with actually putting bits onto the communication
medium. This lesson will give you detailed descriptions of various compo- Lesson Time:
nents of a computer network that correspond to the Physical layer of the 1 hour
OSI model.

at
Objectives

y
To understand the components of a computer network that correspond to the

op
Physical layer of the OSI Model, you will:
eM
4A Examine bounded data-transmission media and connectors.
You’ll examine network cabling such as twisted-pair, coax, and fiber optic

tC
cables. You’ll also examine the advantages and disadvantages of each and
determine when you should use them.

4B Examine unbounded data-transmission media and connectors.

No
You’ll examine unbounded media including radio, microwave, and laser
c

transmissions. You’ll also compare these to some of the bounded trans-

mission media.
en

4C Distinguish between analog and digital signaling.

Do

This topic covers use of switches to represent digital values (binary val-
ues) and differentiate between digital and analog devices.

4D Compare baseband and broadband transmissions.

You’ll examine these two signaling techniques and identify where each is
se
r

used.
fe
ea
Pl
Re

Lesson 4: The Physical Media 69

 TOPIC 4A
Data-transmission Media
For computers to communicate, they need to be connected. The data-transmission

ial
media is the channel or conduit through which data is transmitted. Computers can
be connected by many different kinds of data-transmission media, which can gen-
erally be divided into two categories: bounded and unbounded. You can think of
bounded media as wires or cables and unbounded media as the airwaves. This
topic introduces the most common bounded media, including twisted-pair cable,
coaxial cable, and fiber optic cable; and unbounded media, including radio, laser,

er
bounded media: and infrared.
A physical product, such as
cable, through which data is You should be aware of the characteristics of each medium when deciding on the
transmitted. best one for your network needs. There are physical limits with each type of
medium. For example, attenuation, or weakening of a signal over distance, is a
factor in many types of transmission media. The amount of data that can be car-

at
ried on any particular transmission medium is referred to as bandwidth.
attenuation:
Degradation of signal. Bandwidth is measured in Hertz (cycles per second) or bits per second (bps).

y
Although there are other considerations in a network’s performance, the general
rule is that the greater the bandwidth, the greater the amount of data that can be

op
eM
transmitted. How fast the data moves within the medium is referred to as the data
rate , and it is also measured in bps. Don’t confuse bandwidth with data rate.
bandwidth:
The difference between the One way to think of it is that bandwidth can be compared to the number of lanes
highest and lowest on a highway and the data rate is the speed for the highway. Bandwidth and data

tC
frequencies used for a rate combined equal throughput. A data-transmission medium with a high band-
communication channel.
width and a low data rate could have the same throughput as a data-transmission
Generally, more bandwidth
means greater transmission medium with a low bandwidth and a high data rate.
capacity.
No
c

data rate:
How fast the data moves
en

within the medium.

Do

throughput:
A measure of productivity on
se

a device or network.
r
fe
ea
Pl
Re

Figure 4-1: Bandwidth versus data rate.

70
Bounded Media
Bounded media transmit signals by sending electricity or light over a cable.

Twisted-pair Cable

ial
In its simplest form, twisted-pair cable (TP) contains a pair of insulated copper
wires that are twisted around each other and encased in a protective sheath. The
following figure illustrates the two major types of twisted-pair cabling: unshielded
twisted-pair (UTP) and shielded twisted-pair (STP). In STP, the inner wires are
encased in a sheath of foil or braided wire mesh.

er
at
y
op
eM
twisted pair:
Cable consisting of at least
two insulated wires that are

tC
intertwined to reduce
electromagnetic interference.
No
Figure 4-2: UTP and STP cables.
c

unshielded twisted pair:

Twisted pair wire that isn’t
protected from EMI through
the use of shielding.
en
Do

shielded twisted pair:

Twisted pair wire that’s
surrounded by a foil or
braided metal sheath for the
se
r

purpose of reducing
susceptibility to
fe

electromagnetic interference.
ea
Pl
Re

Lesson 4: The Physical Media 71

 In both UTP and STP, the wires are twisted around one another to reduce electro-
magnetic interference (EMI). EMI can occur as noise (interference from outside
sources such as power sources) or as crosstalk (signal overflow from an adjacent
wire), and can severely hinder data communications. The shielding of STP is
EMI: designed to further reduce its susceptibility to interference.
Electromagnetic interference.
Noise or crosstalk that can The number of pairs of wires contained in a twisted-pair cable varies. Also, there

ial
severely hinder data are standards governing how many twists are required per foot of cable. Most
communication. twisted-pair cable is relatively inexpensive. Twisted-pair cable has been used for
years in the telephone industry, so the technology and standards for this type of
cable are well established. However, you should be aware that many standard
telephone cables aren’t twisted pair and are inadequate for most network
crosstalk: applications.
Signal overflow from an

er
adjacent wire. The Electronics Industries Association and the Telecommunications Industries
Association’s (EIA/TIA) 568 Commercial Building Wiring Standard defines five
categories of UTP, which are described in the following table.

Category Description

at
Category 1 Traditional UTP telephone cable. Category 1 can transmit voice signals but not
data. Most telephone cable installed prior to 1983 is Category 1.

y
Category 2 UTP cable made up of four twisted-pair wires, certified for transmitting data up to
4 Mbps (megabits per second).

op
Category 3
eM UTP cable made up of four twisted-pair wires, each twisted three times per foot.
Category 3 is certified to transmit data up to 10 Mbps.
Category 4 UTP cable made up of four twisted-pair wires, certified to transmit data up to 16
Mbps.

tC
Category 5 UTP cable made up of four twisted-pair wires, certified to transmit data up to 100
Mbps.
No
c

The IBM Cabling System

In 1984, IBM introduced its own cabling system, complete with standards and
en

specifications. IBM developed custom cable connectors, face plates, distribution

Do

panels, and more. Much of this specification was designed to support IBM’s net-
working technology, Token Ring.

IBM type Standard label Description

AWG:
Type 1 Shielded twisted pair Two pairs of 22 AWG (American Wire Gauge standard*)
se

A measurement of wire
r

diameter, where the diameter wires surrounded by braided wire shield.

of the wire increases as the Type 2 Voice and data cable Two pairs of 22 AWG wires surrounded by braided shield
fe

AWG number gets smaller. used for data surrounded by four additional pairs of 26 AWG
ea

Thus, AWG 22 wire is thicker cables used for voice.

than AWG 26 wire.
Type 3 Voice grade cable Four, solid, unshielded twisted-pair 22 or 24 AWG wires.
Type 4 Not defined
Pl

Type 5 Fiber optic cable Two 62.5/125 micron multimode optical fibers.
Re

Type 6 Data patch cable Two pairs of 26 AWG stranded wires with a dual foil and
braided shield.
Type 7 Not defined
Type 8 Carpet cable Two shielded twisted pairs of 26 AWG cables housed in a
flat jacket (designed to be run under carpets).

72
IBM type Standard label Description
Type 9 Plenum cable Two shielded twisted-pair cables in a fire-safe jacket.

* AWG is a measurement of wire diameter, where the diameter of the wire

increases as the AWG number gets smaller. Thus, AWG 22 wire is thicker than

ial
AWG 26 wire.

Twisted-pair Connectors
A common connector type used with twisted-pair cable is the RJ-45 connector.

er
(Certainly, other types of connectors are used, though less commonly than the
RJ-45.) The RJ-45 connector is similar to a typical telephone plug (RJ-11); how-
ever, the RJ-45 is larger. It contains eight cable connections compared to the
RJ-11, which contains only four.
RJ-45:
A connector used with

at
twisted-pair cable that
contains eight cable
connections.

y
op
eM
tC
Figure 4-3: RJ-45 connectors.
Additionally, large twisted-pair networks can include the following hardware
components:
No
Component Purpose
c

Distribution rack and rack Used to create more room for cables where floor space is at a
shelves premium, and to centralize and organize network connections.
Expandable patch panel Support for multiple ports. Transmission speeds of 100 Mbps.
en
Do

Jack couplers (patch cord) Single or double RJ-45 jacks that snap into patch panels, and
wallplates to connect two ports, such as the computer to the
wallplate. They support data rates to 100 Mbps.
Wall plates Supports two or more jack couplers.
se
r
fe
ea
Pl
Re

Lesson 4: The Physical Media 73

 ial
er
Figure 4-4: Twisted-pair patch panel.
STP is capable of handling more data with greater speed than is UTP. Twisted-
pair cable is susceptible to eavesdropping; however, STP is less susceptible than

at
UTP. The following table lists the advantages and disadvantages of twisted-pair
cabling.

y
Advantages Disadvantages

op
Is inexpensive.
eM
Has well-established products
Is susceptible to EMI and eavesdropping.
Has a low bandwidth.
and technology.
Is easy to install.

Crimping RJ-45 Connections

tC
No
c

To create an RJ-45 modular cable and properly terminate it, you’ll most likely
use the T568A wiring scheme. This scheme was created by the EIA/TIA (Elec-
tronic Industry Association/Telecommunications Industry Association). T568A
en

uses 8 wire twisted-pair cabling with the following wiring scheme:

Do

Pin Wire color Ethernet Token Ring

1 White/Green Transmit+ ----
2 Green Transmit- Ground
3 White/Orange Receive+ Transmit+
se
r

4 Blue --- Receive+

5 White/Blue --- Receive-
fe

6 Orange Receive- Recovery

ea

7 White/Brown --- Ground

8 Brown --- ---
Pl
Re

74
The T568A wiring scheme can be used for both 10BaseT and Token Ring
networks. To make an RJ-45 modular cable:
1. Slit the PVC covering on the cable about 1.5 cm from the end, being careful
not to cut or damage any of the internal wires.
2. Spread out the wires in the order shown in the T568A table, keeping them
parallel and close together.

ial
3. Clip the wires so they are even in length.
4. Slide the wires into the modular plug with the White/Green wire on the left.
Make sure the wires are all the way in under the metal teeth in the plug.
5. Insert the plug into the crimper tool while not getting any slack in the wires.
Close the handles on the crimper hard so that all of the teeth in the plug bite

er
through the wires.
6. Check the plug by looking to see if you see any wires that weren’t bitten.
Gently pull on the cable to see that it’s firmly crimped. You can also use a
cable tester to check continuity.

at
Coaxial Cable

y
Coaxial cable, most often called coax , contains a single-wire inner conductor

op
eM
surrounded by a layer of insulation, a shield of braided wire mesh or foil, and an
outer layer of insulation. The inner conductor carries the data and can be either
solid (usually copper) or stranded. The following figure illustrates typical coax
cable.

tC
coaxial cable:
A data-transmission medium
that contains a single-wire
inner conductor surrounded
by a layer of insulation, a
No
shield of braided wire mesh
c

or foil, and an outer layer of

insulation.
en

Figure 4-5: Coaxial cable.

Do

The purpose of the shield, typically a braided wire mesh, is to act as a ground to
protect data from EMI. If the core and the shield touch, the cable will short, and
noise or stray data signals will flow onto the core, destroying the data signal
being transmitted. Thus, there must be a layer of insulation between the core and
se

the shield. The entire cable is wrapped in another layer of insulating material,
r

usually made of rubber, Teflon, or plastic. Coax gets its name because of the fact
that the conductor and the shield share the same axis, thus are ″co-axial.″
fe
ea

There are two types of coaxial cable: thin (thinnet) and thick (thicknet) as shown
in the following figure.
Pl
Re

Lesson 4: The Physical Media 75

 ial
er
Figure 4-6: Thicknet versus thinnet coaxial cable.
The following table provides a comparison of thinnet and thicknet coaxial cable.

Thinnet coaxial cable Thicknet coaxial cable

at
Can carry a signal up to approximately 185 Can carry a signal up to approximately 500 meters
meters (607 feet). (1,640 feet).
Is about 0.25 inches thick. Is about 0.50 inches thick.

y
Is flexible and easy to work with in connecting Is thicker than thinnet and is used mainly as a
network computers. ″backbone″ to connect thinnet networks together.

op
eM
Coax cable can be further categorized based on a number of specifications:

tC
Coax cable type Impedance Description
The RG stands for Radio
Grade and is an old term RG-58/U 53.5 ohms Solid copper core thinnet cable.
defined by the military RG-58 A/U 50 ohms Stranded copper core thinnet cable.
specifications that originally
No
RG-58 C/U 50 ohms The military version (MILSPEC) of RG-58 A/U.
designated these
c

specifications. RG-8 50 ohms Stranded core thicknet (broadband) cable.

RG-11 75 ohms Dual-shielded, solid core thicknet cable.
RG-59 75 ohms Dual-shielded, solid core cable used for broadband
en

transmissions, such as cable television. Not suitable for

Do

For more information on most networking applications.

these and other cable
RG-6 75 ohms Similar to RG-59, but larger in diameter. Can handle
specifications, visit the
Belden Web site at www. higher bandwidths than RG-59.
belden.com. RG-62 93 ohms ARCnet networking cable, not suitable for Ethernet
networking.
se
r
fe
ea
Pl
Re

76
Coaxial Connectors
A British Naval Connector (BNC)—also called a Bayonet Nut Connector, Bayonet
Neill Concelman, or Bayonet Navy Connector —is usually used for thinnet
coaxial cable. The connector fits into a coupling and is twisted to lock it into
place. A BNC T-connector provides three-way connection: two connections for
the network and one connection for the transceiver device. At the end of a length

ial
of cable, you will often see terminators. A terminator is a resistor attached to the
end of the cable. Its purpose is to prevent signal reflections, effectively making
the cable ″look″ infinitely long to the signals being sent across it. The thinnet BNC:
connector types are shown in the following figure. (British Naval Connector or
Bayonet Nut Connector)
Connector usually used for
thinnet coaxial cable.

er
Provides three-way
connection—two
connections for the network
and one for the transceiver
device.

at
terminator:

y
A resistor attached to the end
of the cable to prevent signal

op
eM reflections, making the cable
appear to be infinitely long
to the signals being sent
across it.

Figure 4-7: Coaxial connection hardware.

tC
No
c

Thicknet coaxial cable uses a vampire tap to connect to a transceiver. Every few
feet of cable is marked to indicate where a new connection can be made. The
tool to make the connection is a metal spike that pierces the cable. A special con-
en

nector is then screwed into the thick coaxial cable.

Do
se
r

Figure 4-8: A vampire tap, ready to be installed on the cable.

fe
ea
Pl
Re

Lesson 4: The Physical Media 77

 A cable connects the transceiver and the Attachment Unit Interface (AUI) port on
the network adapter. An AUI connector is a 15-pin D-shaped connector (a DB-15
connector) that looks like a parallel port connector. Another commonly used name
for an AUI connector is a DIX connector, named for the three companies that
AUI connector: developed it—Digital, Intel, and Xerox.
(Attachment Unit Interface) A
15-pin D-shaped connector Most specifications and many local electrical codes require you to ground one

ial
(a DB-15 connector) used in point on a coax segment. Such a ground connection allows static electricity to
thicknet cable networks. harmlessly drain off rather than build to the point where people or equipment
could be harmed. Make sure you ground your cable at only one point. Multiple
grounds on an Ethernet segment will cause network errors, and could increase the
risk of damage to equipment or injury to people.
DIX connector:
Another name for an AUI The following table lists the advantages and disadvantages of coaxial cable:

er
connector, named for the
three companies that Advantages Disadvantages
developed it: Digital, Intel,
and Xerox. Is light, flexible, and easy to work with. Can be difficult and expensive to install.
Is not as susceptible to EMI and Is susceptible to EMI and eavesdropping, although

at
eavesdropping as twisted-pair cabling. better than twisted-pair cable.
Is inexpensive. Is more expensive than twisted-pair cabling.
Has well-established products and technology.

y
Is more resistant to attenuation (the weakening
of a signal over a distance); is better than

op
eM
twisted-pair cabling for longer distances.

Crimping BNC Connectors on Coaxial Cables

tC
If you are using thinnet, you’ll need to either purchase pre-made cables, or create
your own. The following steps explain how to terminate BNC connections:
No
1. Start with RG58 coaxial cable, a crimp, pin, and BNC connector.
c

2. Slide the crimp over the coax cable, then using wire strippers, strip off a
length of the outer covering that is slightly shorter than the length of the
en

crimp.
Do

3. Pull the braided layer back over the stripped cable, exposing the insulation
around the center conductor. Be sure not to cut off the braiding.
4. Remove a length of insulator the same length as the bottom of the pin to the
indentation on the pin, then insert the center conductor all the way into the
pin.
se
r

5. Using the crimping tool, crimp the pin to the center conductor, making sure
it is held tightly.
fe
ea

6. Push the BNC down over the pin until it grabs the cable.
7. Slide the crimp up and over the braid and base of the BNC.
8. Crimp the crimp to the cable with the crimping tool.
Pl
Re

78
Fiber Optic Cable
Fiber optic cable transmits signals of light though a very fine strand of plastic or
glass (optical fiber), called the core . The core is encased in a tube that is made
of glass, called cladding . Light bounces from side to side as it moves through
the optical fiber. The cladding is designed to reflect light back into the optical
fiber. The cladding is then surrounded by a tough outer sheath, called a jacket ,

ial
which provides a protective covering for the optical fiber. Each optic fiber core
can transmit data in only one direction, so a cable generally consists of two
strands in separate jackets. One strand would send data, the other strand would fiber optic cable:
receive it. A cable that transmits
signals of light through a
very fine strand of plastic or
glass (optical fiber) called

er
the core, which is encased in
a tube made of glass, called
cladding, which in turn, is
encased in a tough outer
sheath called a jacket.

at
Figure 4-9: Fiber optic cable.

y
op
eM
Fiber Optic Connectors
Fiber optic connectors are used for fiber optic cable. SMA connectors have a

tC
threaded nut that is twisted to lock it into place. ST connectors use a bayonet
locking system.
The following table lists the advantages and disadvantages of fiber optic cable:
No
Advantages Disadvantages
c

Is nearly impossible to tap, making it secure. Is expensive.

Doesn’t produce (and isn’t affected by) EMI. Is more complex to install than other types of
cable.
en
Do

Has a high capacity for transmitting data, generally Has less-established products and technology
100 Mbps or faster (lab experiments have when compared to other bounded media.
demonstrated throughputs of 1 Gbps).
Is lightweight. Is brittle and breaks easily.
Has very little signal attenuation, enabling data
se

transmission for miles.

r

Uses less energy to operate than other transmission

media.
fe
ea

Cable Grades
Pl
Re

Cables are also graded for fire safety. Many cables contain polyvinyl chloride
(PVC), a plastic used to make the insulation in the cables. PVC is flexible, mak-
ing cable made with it easy to install. However, when PVC burns, it creates
poisonous gases. National and local fire codes regulate where PVC cables can

Lesson 4: The Physical Media 79

 exist in a building. One place they cannot run is within a plenum. A plenum is an
air-handling space (part of the heating/cooling system) and is often a convenient
place to run cables.
plenum cable:
Cable made of special Because poisonous gases and flames can spread quickly throughout the building
materials in the insulation via a plenum, a special cable must be used. Plenum cable, which gets its name
layers that make it fire from the plenum space, has special materials in the insulation layers that make it

ial
resistant. When it burns, it fire resistant. When plenum cable burns, it produces a minimal amount of smoke
produces a minimal amount and chemical fumes.
of smoke and chemical
fumes.

er
at
y
op
eM
tC
No
c
en

Figure 4-10: Examples of locations for PVC and plenum cables.

Do

Plenum cable was designed to meet standards set by the National Fire Protection
Agency (NFPA) and others for smoke and toxic vapors in air-handling spaces.
Local fire codes have also affected the design of plenum cabling, requiring that
the cables themselves cannot be a conduit for fire. The special materials in the
se

insulation also make plenum cabling more expensive and less flexible than PVC
r

cable.
fe

Cable rating Standard for performance on a flamespread test

ea

FT4 Flame Test 4 rating as described by the Canadian Standards Association (CSA)
standard C22.2.0.3-1992.
FT6 Flame Test 6 rating described by the National Fire Protection Agency (NFPA)
Pl
Re

standard 262-1985.
UL-910 National Electrical Code (NEC) standard based on Underwriters Laboratories
(UL) tests for smoke and flame spread.

80
Cable classification Description
CMS, CL2X Restricted cable; must be enclosed in a conduit; may have only up to
10 feet of cable exposed; must pass UL test 1581 VW-1.
CM, CL2 General-purpose cable; used in areas other than risers (a vertical shaft
used to route cabling between floors) or plenums; must pass UL
vertical tray test 1581.

ial
CMR, CL2R Riser cable; used in vertical shafts; must pass UL test 1666.
CMP, CL2P Plenum cable; used in plenum areas; must pass UL test 910 for
smoke and flame spread.

er
Comparing Media Types
The data-transmission medium (or media) that a network uses determines many
of the characteristics and limitations of the network. The following are some of
the things you should consider when you compare data-transmission media:

at
• Susceptibility to EMI . EMI can come from a variety of sources, including
fluorescent lights, elevators, water coolers, power supplies and transformers,

y
high-voltage equipment (such as welders), and components of other commu-
nications systems (such as radio transmitters). EMI can severely hinder data

op
communications.
eM
• Security . Not only are many types of data-transmission media susceptible to
EMI, many emit EMI as well. This can create a problem in crowded wiring

tC
closets, where cables can interfere with each other or create a security
problem. A nearby receiver could pick up electromagnetic signals emanating
from a cable. With the proper equipment, those signals could be decoded and
interpreted (wiretapped).
•
No
Bandwidth, or the transmission capacity of the medium . Typically, the
c

greater the bandwidth, the greater the amount of data that can be transferred
over a network.
• Cost to purchase and maintain . Although the initial cost might be low for
en

certain media, you should also consider other expenses that you will experi-
Do

ence over the life of the medium, such as maintenance and upgrade costs.
Also, certain products are more durable than others and might hold up to the
ravages of time, use, and the environment better than others.
• Compatibility with other products in your network . If possible, you should
purchase products that adhere to established standards. It is more likely that
se
r

future networking products will be compatible with standard media than

non-standard.
fe

• Ease of troubleshooting . You should consider the significance of your

ea

network. If it fails, what are the costs of extended down time due to diffi-
culty in troubleshooting. What is the level of expertise in your support
personnel? Do they have the knowledge or experience to troubleshoot a
Pl

more complex physical media?

Re

Lesson 4: The Physical Media 81

 TASK 4A-1:
Matching the cable type with its properties
Objective: For each of the following statements, identify which transmis-
sion media is being described.

ial
1. Can’t be tapped and result in stolen data.

2. A flexible cable that is easily installed, but shouldn’t be placed in crawl

spaces.

er
3. Better for very high-speed, high-capacity data transmission because of
the lack of attenuation and the purity of the signal.

at
4. The most commonly-used cable for existing telephone systems; already

y
installed in many office buildings.

op
eM
5. Uses a foil wrap between and around wire pairs; uses internal twisting
of the pairs to give improved protection from EMI.

APPLY YOUR KNOWLEDGE 4-1

tC
No
Suggested time:
c

15 minutes
Viewing videos about cables and connectors
You’ll use the Certification CD included with your manual to view digital video
en

presentations about cables and connectors.

Do

1. Load the Digital Video component of the Network+ Certification CD-ROM.

2. From the Course list, choose Network Cables And Devices.

se

3. Click on Cables and Connectors, and then click Overview to begin the video
r

presentation. When the presentation has finished, you’ll be back at the topic
list, ready to pick out your next video.
fe
ea

4. Click Crimping RJ-45 And BNC Connectors, and then click Overview to
begin the video presentation.

5. Return to the Main Menu, and exit the Network+ Certification CD-ROM.
Pl
Re

82
Building a Network
Let’s now examine some physical networking components. The following
optional activity lets you find the appropriate network components that go
together to form a LAN.

ial
OPTIONAL TASK 4A-2:
Building a network
Objective: To build a network without any computers (just the network-

er
ing components that would be required to create a network).
Setup: You can complete this task if your instructor has provided net-
work cards, cabling, and any other devices which would be
required to cable together two or more computers in a LAN.

at
1. Identify the network card and associated cable type for your network.

2. Connect the network card and cabling, using any other components you

y
might need.

op
3.
eM
Explain to your classmates what kind of network this would be used for
and what components are included in your network.

4. If other topologies are available, return the components you already used,

tC
and then connect another topology’s components together.

5. Again, explain what kind of network you’ve created and what compo-
nents were included.
No
c

6. Return all of the components.

TOPIC 4B
en
Do

Unbounded Media
Unbounded media transmits data without the benefit of a conduit—it might trans-
se
r

mit data through open air, water, or even a vacuum. In addition to various forms
of radio, unbounded media includes links based on laser, infrared, and other light
forms. Unbounded media, such as infrared, can be used to connect computers or
fe

other devices to a network without a cable, creating a wireless component to your

ea

cabled network.
A wireless network enables a business to:
•
Pl

Provide users with temporary connections to an existing LAN.

Re

• Create a backup to an existing LAN.

• Provide a certain degree of portability to its network, allowing users to work
from remote locations.
• Extend its networks beyond the physical limitations of cabling.

Lesson 4: The Physical Media 83

 • Enable employees, such as salespeople or doctors, to move around
frequently.
• Provide network connections in buildings where cabling would be difficult,
such as buildings designated as historical landmarks.
• Provide network connections in busy areas, such as lobbies, where cables
could be unsightly or even dangerous (people could trip over cables, but are

ial
not likely to trip over radio waves).
Other unbounded media, such as microwave, can be used to connect networks
over very long distances. Typically, unbounded media are more prone to eaves-
dropping and electromagnetic and atmospheric interference (such as rain and fog)
than are bounded media.

er
Radio
Although the term radio can refer to a large portion of the electromagnetic spec-

at
trum, we will refer to the range of frequencies that includes FM radio, television
(VHF and UHF), and computer networks. Transmissions in this frequency range
are limited by distance because they don’t bounce off the ionosphere and aren’t

y
as affected by atmospheric conditions (unlike microwave transmissions). For these
reasons, radio is suitable for data communications. However, because transmis-

op
eM
sion frequencies of radio transmissions tend to be lower than those of microwave
transmissions, data transmissions are constrained by physical objects (like build-
ings and trees). Also, radio transmission systems tend to have slower data rates
than some other types of unbounded media.

tC
There are two types of radio transmissions: narrow-band (single-frequency) radio
and spread-spectrum radio. In narrow-band radio, you tune both the transmitter
and the receiver to a certain frequency. Just as you can receive a music radio sta-
tion inside a building, data transmissions using radio frequencies can pass through
No
narrow-band radio: walls. This means a radio data link doesn’t require line of sight focusing. Typical
c

A radio signal where both radio network devices have a broadcast range of about 5,000 square feet.
the transmitter and receiver
are tuned to a certain Because radio systems transmit at a high frequency, transmissions cannot occur
frequency to send and through steel or load-bearing walls. (Just as you can hear more static than music
en

receive data. inside buildings built with walls of this type.)

Do

Spread-spectrum radio works differently than broadcast radio. While there are two
primary types of spread spectrum radio, by far the most common system is
spread-spectrum radio: frequency-hopping spread spectrum . In this type of system, data is transmitted
A radio signal where the over a range of frequencies, divided into channels , or hops . The transmitter and
se

transmitter and receiver tune the receiver tune into a specific channel for a predetermined period of time; they
r

into a specific channel for a then switch to a different channel, and so on, continuing this pattern until the data
predetermined period of
has been transmitted and received. Eavesdropping is discouraged because only the
fe

time. They then switch to a

sender and receiver know the frequencies selected (out of the millions possible),
ea

different channel periodically

until the data has been the frequency switching order, and each channel’s transmission times. In general,
transmitted and received. spread-spectrum radio is a slow transmission method compared to other
unbounded media.
Pl
Re

84
Typically, frequency-hopping spread spectrum can offer 250 Kbps throughput,
though some implementations can reach 4 Mbps or faster. The range of this sys-
tem is typically two miles outdoors and about 800 feet indoors.
The second type of spread spectrum radio, Direct Sequence Spread Spectrum ,
spreads the signal across many frequencies simultaneously. DSSS combines a Additional information on
radio carrier signal with the signal in such a way that the resulting output is a frequency-hopping and

ial
wide-bandwidth, low-power wave. Signals generated with this technique appear DSSS spread-spectrum radio
as noise to eavesdroppers. This system is complex and expensive. It is sometimes is available at http://www.
used by the military because intercepting this type of transmission is very cwt.vt.edu/faq/ss.htm.
difficult. Range and throughput are limited with this type of system.

Note: You must have a license to use some bands of radio as a network medium,

er
and it is susceptible to EMI and eavesdropping. Many products use specific fre-
quencies set aside by the Federal Communications Commission (FCC) for
unlicensed use. These bands are called Industrial, Scientific, and Medical (ISM)
bands. Just because you might not need a license to use these bands does not
mean that you can pick any radio frequency within these designated ranges. The
FCC regulates specific bands and what they can be used for, even if the FCC

at
doesn’t require a license for such use.

Cellular phone service (an example of radio communications), which uses the

y
ISM bands, can provide an efficient portable network connection. Cellular Digital

op
Packet Data (CDPD) technology enables packet (network style) communications
over the cellular telephone voice network.
eM
tC
Terrestrial Microwave
Terrestrial microwave is commonly used for long-distance voice and video trans-
missions, and for short-distance high-speed links between buildings. A common
No
type of microwave antenna is a parabolic dish about 10 feet in diameter. In an
area of rough terrain, microwave technology is often less expensive than cabling.
c
en
Do
se
r
fe
ea
Pl
Re

Figure 4-11: Terrestrial microwave.

Lesson 4: The Physical Media 85

 Microwave is similar to radio, but a major difference is that radio broadcasts tend
to be omnidirectional (broadcast in all directions), while microwave transmissions
are focused in one direction (to the other dish). For this reason, microwave anten-
nas must be aligned precisely for successful transmission. Transmission from far
beyond the horizon cannot be picked up. The atmosphere bends microwave sig-
nals toward the earth somewhat, extending the limit of microwave transmissions
beyond the line of sight, although not enough to follow the curvature of the

ial
earth’s surface perfectly. The following table shows how elevating the antennas
above the ground can extend the distance between points. The measurements in
feet and miles are approximate.

Elevation of antennas Maximum distance between points

er
10 meters (32 feet) 26.07 kilometers (16 miles)
100 meters (328 feet) 82.45 kilometers (51 miles)
500 meters (1,640 feet) 184.35 kilometers (114 miles)

High frequencies generally correspond to a high capacity for carrying data, so

at
very high capacities are available within the high end of the microwave range.
However, higher frequencies are more susceptible to atmospheric interference like
rain. Also, buildings, motor vehicles, mountains, and other solid objects can

y
obstruct microwave transmissions. Again, it is important to note that a license is

op
required for microwave transmissions and the transmissions are susceptible to
eavesdropping.
eM
tC
Satellite Microwave
To enable networks to traverse distances greater than those provided by terrestrial
microwave, communications satellites high above the earth’s surface can relay
signals between ground stations. When a ground station transmits data to the
No
communications satellite (a process called uplinking ), the satellite relays the data
c

transmission to another ground station (a process called downlinking ), and then

repeats the data. The ground station can be mobile. Typical communications satel-
lites are in a geosynchronous orbit—their path follows the rotation of the earth,
en

enabling them to remain in the same relative location above the earth.
Do
se
r
fe
ea
Pl
Re

86
 ial
er
at
y
op
eM
Figure 4-12: Satellite microwave.
The physics involved dictate that communications satellites must be 22,300 miles

tC
above the earth. Microwave signals move data at the speed of light. Even at this
high speed, the great distance (more than 44,600 miles by the time a data trans-
mission returns to earth) involved in a satellite microwave transmission means
that there is a significant delay between the time data is sent and the time they
No
are received. When combined with the time it takes for the satellite to repeat the
c

signal, the delay is typically 240 to 300 milliseconds. Sometimes you can notice
this delay in long-distance telephone conversations that use satellite microwave
transmissions; the echo of your own voice is slightly delayed. For some network-
en

ing applications, this delay can introduce problems. Satellite microwave

Do

technology is expensive, requires a license, is susceptible to eavesdropping, and is

affected by weather conditions.

Laser
se
r

A laser transmitter uses a highly focused beam of light pulses to transmit data.
The receiver uses a photoelectronic (light-sensitive) device to convert the laser
fe

back into the original data. Like optical fiber, laser uses light, which occurs at a
ea

higher frequency than the frequencies used in microwave and various forms of
radio. Because of this, laser is capable of very high data rates. However, laser is
also susceptible to attenuation (loss of power) and interference, which weakens
Pl

the signal.
Re

Lesson 4: The Physical Media 87

 One popular type of laser networking system is point-to-point infrared. Although
the interface devices are expensive, they are resistant to eavesdropping, and
capable of high data transmission rates. However, laser transmissions are affected
by atmospheric conditions, susceptible to high-intensity light, and require line-of-
sight positioning. Laser (not infrared) is commonly used for building-to-building
links and can reach typical LAN speeds or beyond. This is a low-cost way to link
two buildings, as compared to digging a trench and laying cable.

ial
er
at
y
op
eM
Figure 4-13: Infrared light.
Infrared transmitters are fairly common devices used in many homes—typical

tC
remote controllers for televisions, videocassette recorders, and stereos use infrared
light to transmit data. Infrared systems don’t require a focused beam, yet aren’t
omnidirectional either. Infrared signals reflect easily off many hard surfaces typi-
cal of an office or home. For example, the television remote control can be aimed
No
at the ceiling and still function properly. The beam is more dispersed, and hence
c

is called broadcast. With this type of system, it is easier to line up the transceiv-
ers because of the broadcast nature of infrared light.
en
Do

TASK 4B-1:
Matching the unbounded transmission medium with
its properties
se
r

Objective: For each of the following statements, identify which

unbounded transmission media is being described.
fe
ea

1. Transmissions can’t occur through steel or load-bearing walls.

Pl
Re

2. Resistant to eavesdropping and capable of high transmission rates; how-

ever, it is susceptible to attenuation and interference.

88
3. Frequencies are divided into channels or hops.

4. Has a transmission delay of typically 240 to 300 milliseconds.

ial
5. Commonly used for long-distance voice and video transmissions, and for
short-distance high-speed links between buildings.

er
A Comparison of Media Types
The following table compares the different data-transmission media:

Twisted-pair Coaxial Fiber optic

at
cabling cable cable Unbounded media
Susceptibility to High Moderate None Low to moderate; radio is
EMI high

y
Security Low Low High Typically very low

op
Bandwidth
(capacity)
High (1-100 Mbps) Moderate
(10 Mbps)
eM
High (100
Mbps)
Varies widely

Cost to purchase Low to moderate Moderate Moderate to Moderate to high

and maintain high

tC
Ease of Easy Easy Difficult Varies widely
installation and
maintenance
Attenuation High Moderate Low Varies; affected by weather
No
and other objects
c
en

TASK 4B-2:
Do

Determining the type of media to use

Objective: For each of the following scenarios, indicate which type of
transmission media you would use, and explain why.
se
r

1. Your offices are located in an historic 19th-century building. You don’t

fe

want unsightly cables running across the floor, but you are not allowed
ea

to run cables through the walls, floors, or ceiling.

Pl
Re

2. You need to provide a high-speed connection between two offices that

are separated by a courtyard. The offices are in separate buildings.

Lesson 4: The Physical Media 89

 3. You need a high speed connection between graphic workstations that are
in the same office. You have no restrictions except the need for speed.

ial
4. You need to run a connection through a tunnel connecting two
buildings. The tunnel contains many high voltage lines and large electric
motors. You are concerned about security.

er
TOPIC 4C
Signaling

at
Signaling is how computers represent data as an electrical or light impulse
pattern. It is a function of the Physical layer of the OSI model and is specific to
the type of network you use. There are two general approaches for representing

y
data: digital signaling and analog signaling . Because of the wide variety of

op
equipment used in computer networks, you are likely to find both schemes used
eM
somewhere in the network:
• Digital devices use discrete values, such as the presence or absence of an
electrical signal, to represent portions of the data.

tC
• Analog devices use a continuous variable signal (a waveform) to represent
portions of the data.
No
Using Switches to Represent Digital Values
c

Most electronic computers use electronic switches to represent digital values. A

single switch can represent the most basic element of information, called a bit. A
en

switch has two states: off and on. The value contained in a bit corresponds to the
Do

state of the switch, 1 or 0, depending on whether the switch is on or off. To rep-

resent a number larger than 1, you have to use more than one switch. A simple
way to represent values larger than 1 is to have as many switches as the highest
number you need to represent. For example, a series of 200 switches can repre-
sent any number between 0 and 200. Even with today’s level of miniaturization,
se
r

this method would require a huge computer to perform simple tasks.

To represent and process numbers, most digital computers use the binary number
fe

system (base 2), which has only the two digits 1 and 0. In comparison, in the
ea

base 10 numbering system there are 10 digits (0 through 9). To represent numbers
greater than the number of digits available, you add another column to the left.
For example, the number greater than 9 is 10; the number greater than 99 is 100;
and so forth. This is the same scheme used in the binary system, except that you
Pl
Re

use only two digits, 1 and 0. Compare the numbers in the chart below to see how
this works:

Base 10 Base 2 (binary) Base 10 Base 2 (binary)

0 0 8 1000

90
Base 10 Base 2 (binary) Base 10 Base 2 (binary)
1 1 9 1001
2 10 10 1010
3 11 11 1011
4 100 12 1100

ial
5 101 13 1101
6 110 14 1110
7 111 15 1111

er
TASK 4C-1:
Working with binary numbers

at
Objective: To review how large decimal numbers are built in order to
gain an understanding of how binary numbers are built.

y
1. Write each of these decimal numbers in the columns in the following
table. The first number has been filled in for you.

op
eM
The Thousands Hundreds
number (103) (102) Tens (101) Ones (100)
4328 = 4 3 2 8

tC
9488 =
101 =
3=
No
90 =
c
en
Do

Using the table above, you can see that large decimal numbers are built
using the digits 0-9 and by placing those digits in columns representing
powers of ten (thus the reason these are called base-10 numbers).
se
r

2. Using the same technique, write each of these decimal numbers as

binary numbers by filling in the digits 0 and 1 into the appropriate
fe

columns. The first number has been completed for you.

ea

Decimal
number Eights (23) Fours (22) Twos (21) Ones (20)
Pl
Re

14 = 1 1 1 0
2=
12 =
7=
4=

Lesson 4: The Physical Media 91

 From the preceding table, you should see that binary numbers are built just

ial
like decimal numbers. Only the digits 0 and 1 are used. The values of the
columns are different; however, they too are powers of their base number. In
this case, each column is a power of 2.

3. How would you write the decimal number 30 using the columns of the
binary system?

er
4. How many places (columns) would you need to write the decimal num-
ber 255 as a binary number?

at
y
op
eM
Characteristics of Digital and Analog Devices
Analog electronic devices use a variable quantity (such as voltage, which is the
amount of electrical force or pressure) to represent a value. Digital devices (such

tC
as personal computers) use bits to represent values. These bits are called discrete
values.
No
c
en
Do
se
r
fe
ea
Pl

Figure 4-14: Analog waves compared to digital waves.

Re

Analog signals carry continuous values and resemble sine waves. Digital signals
carry discrete values and appear as square waves.

92
TASK 4C-2:
Identifying analog and digital devices
Objective: To determine the difference between analog and digital dis-
plays of information.

ial
1. Identify whether the following devices have analog or digital displays of
information:

er
at
y
op
eM
tC
No
c
en
Do

Figure 4-15: Digital and analog devices.

2. List at least one analog device that is not displayed above.

se
r
fe

3. List at least one digital device that is not displayed above.

ea
Pl
Re

Lesson 4: The Physical Media 93

 TOPIC 4D
Signal Transmissions
Two techniques can be used to transmit signals over the data transmission media.

ial
These are baseband and broadband transmissions. A baseband transmission line
provides only one channel per line, while a broadband transmission line provides
multiple communication channels. An example of a broadband transmission is
cable television, which uses a single coaxial cable to bring many television chan-
nels into a home.

er
at
y
op
eM
tC
No
c

Figure 4-16: Baseband and broadband signal transmission.

en
Do

Baseband
Baseband systems use digital encoding techniques to carry digital data over a
baseband: digital transmission line. Most typically, the presence of a voltage is used to rep-
A signal transmission that resent a bit with the absence of that voltage being a ″space″ between a bit.
uses all of a channel’s Because digital transmissions are affected significantly by attenuation, baseband
se
r

bandwidth. transmission lines are generally limited in length only to a kilometer or so.
In baseband transmission, all of the available frequencies in the transmission
fe

medium are used by one signal. Therefore, the medium provides only one com-
ea

munication channel, over which only one computer can successfully transmit at a
time. Baseband transmission is like two-way radio transmission: users must take
turns talking. In baseband transmission, when one computer transmits, all other
computers must be ″quiet,″ or a collision will occur and the data will be garbled.
Pl
Re

94
Broadband
Broadband systems use modulation techniques to transmit digital data over analog
carrier waves. The amplitude or frequency of the signal can be used to represent broadband:
bits, as can far more complex modulation techniques. By dividing the frequencies A transmission channel that
provided by the medium, multiple communication channels can be provided by a can carry multiple signals
single transmission line. Because broadband systems use analog signaling (which simultaneously.

ial
is less susceptible to attenuation than digital signaling), broadband transmission
lines can be much longer than baseband lines.
Broadband media are unidirectional. Unlike baseband, signals move only in one
″downstream″ direction. To enable devices to transmit to ″upstream″ computers,
there must be two data paths, one through which computers can transmit data,

er
and one through which computers can receive messages. A headend device at the
end of the transmit line forwards data to the receiver line, so that all computers
are effectively downstream of transmitted data.

at
y
op
eM
tC
No
c
en
Do

Figure 4-17: Two types of broadband systems.

As you can see in the preceding figure, dual-cable systems use two cables for the
transmit and receive data paths, while split systems use one cable with two differ-
se
r

ent frequencies for the transmit and receive data paths. The split system offers the
benefit of using a single cable, but the headend in the dual-cable system is much
fe

less complex (and usually less expensive) than the split-system headend. The
ea

split-system’s headend must perform more processing than the dual-cable system,
so the split system’s headend is called an active headend . The dual-cable sys-
tem’s headend is called a passive headend .
Pl
Re

Lesson 4: The Physical Media 95

 TASK 4D-1:
Comparing baseband and broadband
Objective: Identify the key differences between the two signal transmis-
sion schemes in the space provided.

ial
1. Baseband

er
2. Broadband

at
y
op
eM
Summary
In this lesson, you’ve examined bounded and unbounded media and the

tC
associated connectors, compared analog and digital signaling, and compared
baseband and broadband transmissions.
No
LESSON 4 REVIEW
c
en

4A Define bounded transmission media and give an example.

Do

4B Define unbounded transmission media and give an example.

se
r
fe
ea

4C What are the two general approaches for representing data? Define
each.
Pl
Re

96
4D Identify and define the two techniques that can be used to transmit sig-
nals over the data transmission media.

ial
er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 4: The Physical Media 97

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
The Data Link Layer LESSON
5

ial
Overview Data Files:
The second layer of the OSI Model is the Data Link layer. This lesson intro- none
duces the responsibilities carried out at this layer. You’ll look in-depth at the
Lesson Time:

er
MAC and LLC subdivisions of the Data Link layer. You’ll also examine
IEEE 802 specifications. 2 hours

Objectives

at
To better understand the responsibilities of the Data Link layer, you will:

y
5A Identify Data Link layer considerations.

op
eM
In this topic, you’ll identify what types of actions occur at the Data Link
layer and identify the sublayers, Media Access Control and Logical Link
Control.

tC
5B Describe the functions of the Media Access Control (MAC) and Logi-
cal Link Control (LLC) portions of the Data Link layer.
You’ll examine how MAC controls contention, polling, and token pass-
ing, and its role in resolving addresses. You’ll also see how the LLC
No
layer passes information using synchronization.
c

5C Examine the IEEE 802 specifications.

It is important that you know the specifications and limitations of various
en

cabling options. This topic introduces you to the IEEE 802 specifications
Do

and focuses on 802.2, 802.3, and 802.5.

se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 99

 TOPIC 5A
Data Link Layer Considerations
The Data Link layer uses the transmission facilities provided by the Physical

ial
layer to transfer data. Whereas the Physical layer operates on signals, the Data
Link layer operates on data. Among the responsibilities of the Data Link layer
are:
• Sharing the media among nodes; therefore, node addressing is important.
• Controlling data flow between the transmitter and receiver.

er
• Packaging data in frames. Frames add special character (bit) sequences at the
beginning or end of data to indicate the beginning and end of a unit of data
transmission, to provide reliability checks, to synchronize the transmitter and
receiver, and to perform other transmission-control functions. Large blocks
of data might need to be split up (fragmented), transmitted in several frames,

at
and reassembled at the receiver.
• Supporting the data transmission medium. Factors include topology and
media-access method.

y
• Detecting and correcting transmission errors from the Physical layer, on the

op
eM
receiving end.
For the Data Link layer to perform its functions, both the sender and receiver
must work in concert. The networking and connection devices associated with the
Data Link layer are bridges, switches, and network interface cards (NICs). Data

tC
Link-specific information, in the form of headers and trailers, enable the Data
Link layer on both ends of the transmission to communicate. Information in the
headers and trailers can consist of addresses (hardware and physical), frame
length, and upper-layer protocol information.
No
c

Among the Data Link layer considerations that you will examine are the ways in
which communication takes place, such as:
• Standards and protocols. For the Data Link layers on both the sender and
en

receiver ends to work together, they must follow the same protocols. Data
Do

Link layer protocols are often divided into two groups. The lower sub-layer
of the Data Link layer, the Media Access Control (MAC), performs Data
Link layer functions related to the Physical layer, such as controlling access
and encoding data into a valid signaling format. The upper sub-layer of the
MAC: Data Link layer, the Logical Link Control (LLC), performs Data Link layer
se

Media Access Control. The

r

functions related to the Network layer, such as providing and maintaining the
portion of the Data Link layer
that controls access to the link to the network. Many of the actual protocols functioning in these sub-
layers will be discussed in a later lesson.
fe

communication channel.
•
ea

Techniques for providing the above-mentioned functions including framing,

flow control, and error checking pertaining to the Physical layer.
LLC:
Pl

Logical Link Control.

Re

Performs Data Link layer

functions related to the
Network layer, such as
providing and maintaining
the link to the network.

100
TASK 5A-1:
Discussing Data Link layer considerations
1. Compare the operation of the Physical layer and the Data Link layer.

ial
2. Which networking and connection devices work at the Data Link layer?

er
3. What are the sublayers of the Data Link layer?

at
TOPIC 5B

y
Media Access Control (MAC) and Logical Link

op
Control (LLC)
eM
The Data Link layer is divided into two sublayers: the Media Access Control

tC
(MAC) layer and on top of that, the Logical Link Control (LLC) layer.

MAC
No
In communication systems where more than one transmitter has access to the
c

same communication channel, a standard procedure must be established to pre-

vent or handle the problem of multiple transmitters trying to send messages out
on the media at the same time. If signals are transmitted simultaneously, colli-
en

sions occur. A collision is the loss of data due to signals mixing together and
Do

becoming damaged beyond effectiveness. The Media Access Control (MAC) sub-
layer contains methods that logical topologies can use to regulate the timing of
data signals and eliminate collisions. collision:
The loss of data due to
The concept behind MAC is similar to a staff meeting in which a variety of dif- signals mixing together and
se

ferent methods can be used to ensure that everyone has an opportunity to speak. becoming damaged beyond
r

Potential problems include: effectiveness.

•
fe

Knowing when a speaker has finished speaking, thus preventing

interruptions.
ea

• Determining who has the microphone when two speakers try to speak at the
same time.
•
Pl

Determining how long a speaker can speak.

Re

• Determining speaking order.

Lesson 5: The Data Link Layer 101

 In network environments, media-access methods compensate for these problems.
Three common media-access methods are contention, polling, and token passing .
When deciding the best method for your network, keep your network’s character-
istics in mind, such as the number of network devices, quantity of transmissions,
and how quickly data needs to reach its destination.

ial
MAC Addresses
Many things on a network, such as software, hardware, and processes, have
addresses. The MAC address concerns a device’s actual physical address, which
is usually designated by the hardware manufacturer. Every device on the network
The first six places in the must have a unique MAC address to ensure proper transmission and reception of

er
MAC address on a NIC data. The MAC layer places this address on frames sent out on the network,
identify the manufacturer of
allowing the device with the matching unique address to receive the frame and
the NIC.
accept the data. ARCnet and LocalTalk LAN boards represent devices whose
MAC addresses aren’t set by the manufacturer. The installer usually sets ARCnet
addresses, and LocalTalk addresses are dynamically assigned.

at
Bridges and LAN switches, which deal with the Data Link layer, specifically fil-
ter or forward traffic based on the MAC addresses. Upon receiving frames, a

y
bridge builds tables that list network devices and the segments they reside on by
reading the MAC information in the frames; you don’t have to manually program

op
eM
the table information.

TASK 5B-1:
Discussing the MAC sublayer
tC
No
1. What is the purpose of the MAC sub-layer?
c

2. What is a collision?
en
Do

3. Describe MAC addresses.

se
r
fe
ea

LLC
Pl
Re

Logical Link Control (LLC) is the second of two portions of the Data Link layer.
The MAC and LLC sub-layers work in tandem to create a complete frame. The
portion of the frame for which LLC is responsible is called a Protocol Data Unit
(LLC PDU or PDU) . When the LLC sub-layer passes the PDU to the MAC sub-
layer for transmission, the MAC sub-layer adds the header and trailer information
that makes it a complete frame.

102
One purpose of LLC is to perform upper Data Link layer functions, such as flow
control and management of connection errors. Another is to provide the means
for transmission synchronization, which ensures that data transmissions are syn-
chronized when they travel from transmitter to receiver.
Because LLC resides in the area between the Network layer and the MAC lower
layers, it provides a standard interface through which the Network layer can com-

ial
municate with any of the lower-layer protocols that LLC supports.

er
at
y
op
eM
Figure 5-1: Communication between the LLC and the Network layer.

tC
Frame Transmission and Synchronization
On a network, bits are arranged into classes, including characters, bytes, octets,
and frames. Bits are synchronized either synchronously or asynchronously. How-
No
ever, they are synchronized at only the bit level. Synchronization must take place
c

at another level to accommodate characters, bytes, octets, and frames.

The methods by which transmissions can be synchronized are asynchronous, syn-
chronous, and isochronous. When deciding which method is best for your
en

network, keep in mind that synchronization can have a great amount of overhead,
Do

and that a method not well suited for a network that often experiences significant asynchronous:
delays in transmissions. A bit synchronization
transmission technique that
uses start and stop bits.

Asynchronous Frame Transmission

se
r

Asynchronous frame transmissions leave clock maintenance to each individual synchronous:

fe

device; transmitting and receiving devices use comparable timing, but their inter- Transmission of a bit stream
ea

nal clocks are not synchronized. A start bit is issued, followed by the frame. The of data where the transmitter
devices take no notice of whether another device has transmitted a frame. Timing and receiver are
errors can occur, but the transmission is not affected because the frames are short. synchronized.
An example of asynchronous frame transmission is a user typing on a computer
Pl
Re

keyboard.
isochronous:
Uses a single network device
for clocking and all other
devices set their internal
clocks to this one device.

Lesson 5: The Data Link Layer 103

 Parity-bit Error Detection
You can implement error-detection by adding a parity bit to every frame. A parity
bit is an additional bit sent with a frame. When the receiver adds up the character
bits and parity bit, it should get an even number if even parity error detection is
used, and an odd number if odd parity is used. If the receiver detects an error, it
can request a retransmission of the bad character.

ial
• If even parity error detection is used, then the transmitter sets the parity bit
to 1 or 0 so that the total number of ″1″ bits in the character, when added to
the parity bit, equals an even number. For example, if the character to be
sent consisted of the bits ″01101011″ (which includes five ″1″ bits), then the
transmitter would set the parity bit to ″1.″ Adding the character bits and the

er
parity bit results in an even number (6).
• If odd parity were used in this example, the receiver would have set the par-
ity bit to ″0″ so that character bits added to the parity bit would equal an
odd number (5).
Although the parity bit can successfully indicate a transmission problem that

at
occurs in a single bit, multiple-bit problems might not be discovered by this
error-detection scheme. For example, a transmission problem that changed the
bits ″01″ to ″10″ would not be detected, because the same number of ″1″ bits are

y
present.

op
eM
Typically, the even parity error-detection scheme is used in asynchronous trans-
mission; the odd, in synchronous transmission.
The transmitter and receiver must use the same parity-bit scheme. Parity-bit error

tC
detection is only about 50% efficient.

Synchronous Frame Transmission

No
c
Synchronous frame transmissions employ both the transmitter and receiver to pro-
vide transmission clocks. They do so in the form of either a separate clock line
(discussed in the Physical layer lesson) or a bit-string or a control character or
characters. Each method provides information specifying which bit begins a
en

frame or byte. The receiver uses that information to determine the point at which
Do

it should accept data and count the bits. Because the transmitter and receiver use
the same clock, timing errors are reduced. Synchronous transmissions also reduce
the overhead when transmitting large amounts of data.
During times when no data is being transmitted, fill bits are used to keep the
se

transmission timing accurate.

r

Synchronous frame transmission often employs an error-detection scheme known

fe

as Cyclic Redundancy Check.

ea
Pl
Re

104
Cyclic Redundancy Check
With the Cyclic Redundancy Check (CRC) error-detection scheme, the transmitter
performs calculations (an algorithm) on the contents of the transmission. The
transmitter attaches the result of these calculations (typically in two or more
bytes) to the data and sends it all to the receiver. The receiver performs the same (CRC) Cyclic Redundancy
calculations on the data and compares the result with the one sent by the Check:

ial
transmitter. If the two results are identical, the transmission is assumed to be An error-detection method
used to detect data
good. If not, the receiver can request a retransmission of the data. CRC is more corruption due to
efficient than parity-bit error detection; it has been said that CRC is 99% efficient. transmission errors.

Isochronous Frame Transmission

er
Isochronous frame transmission uses a single network device for clocking. All
other devices set their internal clocks to this one device, which provides continu-
ous clocking. Network devices monitor the transmission for available space
(slots) by using their media-access rules; they can insert data when there is a free

at
slot.
Unlike asynchronous systems, which provide a clock signal for every frame, and

y
synchronous systems, in which the transmitter sends a clock signal with each bit
string or character group, an isochronous system relies only on one network

op
eM
device to keep all network device clocks synchronized.
Advantages of isochronous transmission include:
• Low overhead.

tC
• Transmission rates are guaranteed; the method is deterministic.
A disadvantage is that a reliable fault-tolerant clocking device is required.
No
TASK 5B-2:
c

Discussing synchronization methods

en
Do

1. What is the difference between synchronization at the Data Link layer

and synchronization at the Physical layer?
se
r

2. In asynchronous frame transmissions, does synchronization occur? What

fe

is an example of an asynchronous frame transmission?

ea
Pl
Re

3. If the parity bit is set to ″1,″ what parity error detection is being used in
the example 11010110? What is the result?

Lesson 5: The Data Link Layer 105

 4. In the previous example, what would the result be if the parity were set
to ″0″?

5. In a CRC detection scheme, what is used to determine whether a trans-

mission is good?

ial
CHECK YOUR SKILLS 5-1

er
Suggested time:
2 minutes
Identifying synchronization methods
Identify the synchronization methods described.

at
1. Relies on only one network device to synchronize all network clocks.

y
op
2.
eM
Sends a clock signal with every bit or character group.

3. Sends a clock signal with every frame.

tC
No
Connection Services Provided by LLC
c

LLC supports the following three types of connections for transmitting data:
• Unacknowledged connectionless service. LLC supports simple point-to-point,
en

multipoint, and broadcast transmission of frames. With this type of service,

Do

LLC does not acknowledge that data has been successfully transferred; upper
layers must provide their own reliability checks. Because unacknowledged
connectionless service does not have to perform reliability checks or main-
tain a connection, it is very fast. For this reason, it is the most commonly
used type of service.
se
r

• Connection-oriented service. LLC establishes and maintains a virtual connec-

tion between two stations. This service performs flow control, sequencing,
fe

and error-recovery functions by using acknowledgments. Once the connec-

ea

tion is established, blocks of data can be transferred between nodes until one
of the nodes terminates the connection.
• Acknowledged connectionless service. LLC combines features of the other
Pl

two services. It is similar to unacknowledged connectionless service, except

Re

that it provides a mechanism through which individual frames can be

acknowledged.
An acknowledgment is simply a message that confirms a frame’s reception.

106
 ial
er
at
y
op
Figure 5-2: Types of services provided by LLC.
eM
tC
TASK 5B-3:
Describing LLC connection types
No
1. Describe the three types of connections provided by LLC.
c
en
Do
se
r

2. Identify how a network that uses unacknowledged connectionless service

fe

might cope with transmission errors.

ea
Pl
Re

Flow-control Techniques
Flow-control techniques ensure that data is not transmitted faster than the receiver
can process it. Two forms of flow control include:

Lesson 5: The Data Link Layer 107

 • Window flow control:
— sliding-window (dynamic)
— static-window
• Stop-and-wait flow control (guaranteed rate)

ial
er
Figure 5-3: Sliding-window flow control.

Sliding-window Flow Control (Dynamic)

at
Some networks, such as those using satellite microwave links or large
internetworks, experience a significant delay between the time a transmission is

y
sent and the time it is received. In such networks, the long delays involved in
waiting for acknowledgment can be unacceptable. To overcome this problem, the

op
eM
sliding-window flow control technique does not acknowledge each frame
individually. Instead, several frames are sent as a stream, the size of which varies.
As the transmitter sends the stream, the receiver’s buffer fills. When the incoming
stream surpasses the receiver’s buffer size, a choke packet is sent back; it tells the

tC
transmitter to slow down. The transmitter complies, but then slowly increases the
stream again until the receiver’s buffer fills and the receiver sends another choke
packet. In this way, the data stream size constantly varies. Sliding-window is also
referred to as dynamic or floating-window flow control.
No
c
en
Do

Figure 5-4: Static window flow control.

se
r

Static-window Flow Control

fe
ea

The protocols that use static-window flow control use a specific window size. For
example, a window size may be equal to seven frames. Every frame in the trans-
mission is assigned a number from 1 to 7. An acknowledgment is sent for each
frame. Another frame cannot be sent until a frame with that same number assign-
Pl
Re

ment has been acknowledged. This scheme allows the maximum number of
outstanding frames to be equal to the window size; in this example, the size of
the window is seven frames.

108
Stop-and-wait Flow Control (Guaranteed)
In protocols that use stop-and-wait flow control, the transmitter sends a frame and
waits for the receiver to acknowledge its receipt. The transmitter cannot send
another frame until the previous frame has been acknowledged.
Stop-and-wait flow control protocols are sometimes called reliable, or guaranteed

ial
flow control, because the transmitter receives acknowledgment when data has
been received at the destination. This type of control is established before trans-
missions are sent.

er
at
Figure 5-5: Stop-and-wait flow control.

y
Error Control

op
eM
In the LLC sub-layer, error control consists of being informed of lost or garbled
frames. The following table describes instances that will invoke LLC error
control.

Instance
Frame fails to reach receiver
Explanation
In a connection-oriented or acknowledged connectionless
service, if a negative acknowledgment (NAK) or no
tC
No
acknowledgment is returned to the transmitter, an error is
c

presumed and retransmission occurs.

Offending frame size A receiver will not accept a frame that is higher than the
maximum size limit or lower than the minimum size limit.
en

Differing checksums If a receiver’s checksum differs from the transmitter’s, an error

Do

results. The receiver can request retransmission.

Capacity limit If a channel or device is holding all that it is capable of, any
subsequent frame transmissions can result in an error.
Interference Interference, such as noise, can garble or lose data, resulting
in a transmission error.
se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 109

 TASK 5B-4:
Discussing flow-control techniques
1. Describe the use of sliding-window flow control.

ial
2. Describe the use of static-window flow control.

er
at
3. Describe the use of stop-and-wait flow control.

y
op
4.
eM
List some instances that will invoke LLC error control.

TOPIC 5C tC
No
IEEE 802 Specifications
c

In 1980, the Institute of Electrical and Electronics Engineers (IEEE) met to

develop a set of standards for network cabling, topologies, and media-access
en

schemes (Physical and Data Link layer standards). They convened in February,
Do

The term IEE 802 1980, so the standards that they developed were designated ″802.″ The IEEE
specifications is derived from assigned Technical Advisory Groups to work on various sections of the 802
the fact that the IEEE
project. Each section is related to a particular domain of data communications, as
committee met in 1980 (80)
in February (2), thus 802. shown here.
se
r

Section Domain covered

IEEE 802.1 Communication between two IEEE 802 stations on different WANs or LANs.
fe

IEEE 802.2 Logical Link Control (LLC) for 802.3, 802.4, 802.5, and 802.6
ea

implementations.
IEEE 802.3 Contention-based standards.
IEEE 802.4 Token-passing standards for bus topologies.
Pl
Re

IEEE 802.5 Token-passing standards for star or ring topologies.

IEEE 802.6 Distributed Queue Dual Bus standard for municipal area networks.
IEEE 802.7 Installation and maintenance standards for broadband networks.
IEEE 802.8 Optical-fiber standards.
IEEE 802.9 Isochronous Ethernet standards for ISDN-integrating voice and data
networks for time-sensitive communication.

110
Section Domain covered
IEEE 802.10 Network security or VLANs.
IEEE 802.11 Wireless LANs.
IEEE 802.12 100Mbps contention-based network.

ial
These standards were adopted by ANSI (American National Standards Institute),
and then later revised by the ISO. They are sometimes known as the IEEE 802.x
series. A relationship between each protocol and OSI model layers exists, as
shown in Figure 5-6. The following topics in this lesson provide details about the
IEEE standards listed in the table above.

er
at
y
op
eM
tC
Figure 5-6: IEEE 802.2 and the OSI model.
IEEE 802.2 defines the Logical Link Control (LLC) standard that performs func-
tions in the upper portion of the Data Link layer, such as flow control and
No
management of connection errors. The lower portion of the Data Link layer
Remember that 802.2 works
c

(Media Access Control, or MAC) is provided through other standards, such as

IEEE 802.5 and IEEE 802.3. at the Data Link layer of the
OSI model.
Because LLC resides in the area between the Network layer and the lower-layer
en

IEEE standards (802.3, 802.4, and 802.5), it provides a standard interface through
Do

which the Network layer can communicate with any of the lower-layer protocols
that LLC supports. IEEE 802.2 identifies a frame’s upper-layer protocols and the
Network layer processes used by the frame’s source and destination by adding
header fields.
IEEE 802.2 provides a large degree of flexibility, making it possible for a wide
se
r

variety of network media to be interconnected on the same network.

fe
ea

802.3 and Ethernet

Ethernet, which was first available in the late 1970s, is still an accepted standard
Pl

because of its respectable bandwidth, economical price, and compatibility with

Re

other network products.

Lesson 5: The Data Link Layer 111

 IEEE 802.3 is an extension of the original Ethernet. Because IEEE 802.3 is a
form of Ethernet, it is often called ″Ethernet″ instead of ″IEEE 802.3.″ However,
IEEE 802.3 includes modifications to the classic Ethernet data-packet structure.
Furthermore, while Ethernet spans the Physical and the Data Link layers, IEEE
802.3 spans the Physical layer and only the lower portion of the Data Link layer;
IEEE 802.2 provides the functions of the upper portion of the Data Link layer.

ial
IEEE 802.3 and the Physical Layer
802.3 specifies several Physical layer configuration options, such as baseband and
broadband transmissions, several types of media and topologies, and data rates.
Each configuration option has the same thing in common—they use the Carrier

er
Sense Multiple Access/Collision Detection (CSMA/CD) media-access method.
802.3 supports five different implementation configurations: 10BASE5; 10BASE2
(sometimes called thin-net or cheaper-net); 1BASE5 (also known as Star LAN);
10BASET; and 10BROAD36.

at
The configuration names indicate the following three IEEE 802.3 characteristics:
• Data rate (megabits per second). The first number in the configuration name.

y
• Type of signaling (baseband or broadband). The middle portion of the name.
• Maximum segment length (rounded to the nearest 100 meters). The last num-

op
eM
ber in the name. ( T indicates that the configuration uses twisted-pair cable.)

tC
IEEE 802.3 Specifications
For comparison, the following table describes various IEEE 802.3 media formats,
including their data rates, topologies, and segment lengths. The maximum dis-
tance between stations is 500 meters.
No
c

Data Rate (in Maximum segment

Medium Mbps) Topology length (in meters)
en

10BASE5 (thick coax) 10 Bus 500

Do

10BASE2 (thin coax) 10 Bus 185

1BASE5 (UTP) 1 Star 250
10BASE-T (UTP) 10 Star 100
10BASE-F (optical fiber) 10 Star 500-2000
100BASE-T (twisted pair) 100 Bus 4000
se
r

10BROAD36 (broadband 10 Bus 3600

coax)
fe
ea
Pl
Re

112
TASK 5C-1:
Choosing an IEEE 802.3 configuration
Objective: Use the IEEE 802.3 specifications table to identify which
IEEE 802.3 configuration you would use in each of the fol-

ial
lowing situations, and explain why.

1. You need to connect several offices in a large industrial complex that

spans more than half a mile.

er
2. You need to install a small LAN in a single office. Your budget limits
how much you can spend.

at
3. You plan to install LAN cabling above the ceiling tiles in your office.

y
There are other electrical lines and fan motors above the ceiling tiles,
and the cables must be run through several tight curves.

op
eM
Collision Detection
IEEE 802.3 employs a contention-based media access method, enabling networktC
devices to transmit data at will. When two devices attempt to transmit at the
No
same time, 802.3’s collision detection process kicks in to stop that from happen-
c

ing, as shown in the following figure.

en
Do
se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 113

 ial
er
at
y
op
eM
Figure 5-7: IEEE 802.3’s collision detection process.

Differences Between IEEE 802.3 and Ethernet

• tC
The following is a summary of the differences between IEEE 802.3 and Ethernet:
Ethernet supports the Physical and Data Link layers, while IEEE 802.3 sup-
ports the Physical layer and only the lower half (MAC sublayer) of the Data
No
Link layer. IEEE 802.2 supports the upper half (LLC) of the Data Link
c

layer.
• Ethernet supports only one data rate, transmission medium, and topology.
IEEE 802.3 supports a variety of data rates, transmission media, and
en
Do

topologies.
If you compare Ethernet specifications with those of IEEE 802.3, you will see
that 10BASE5 is the IEEE 802.3 configuration that is most like Ethernet.

Ethernet specifications
se
r

Data Rate (in megabits/second) 10

Maximum distance between stations (in meters) 500
fe

Maximum segment length (in meters) 500

ea

Type of media Thick or thin coax, UTP, and duplex optical fiber
Media Access Control Contention-based CSMA/CD
Topology Bus
Pl
Re

Signaling technique (baseband) Manchester

114
Ethernet Media Types
Next, let’s look at each of the Ethernet media types in more detail.

Thick Coax Cable

ial
10BASE5, also called Thicknet or Thick Ethernet, uses thick, coaxial cable. As in
other types of networks, 802.3 devices attach to the network through network
adapters. 802.3 network adapters typically have an onboard 802.3 transceiver and
use an 802.3 T-connector to tap into 802.3 thin Ethernet cable. Alternatively, they
use an 802.3 Attachment Unit Interface (AUI) cable to attach to a separate trans-
ceiver called an 802.3 Medium Attachment Unit (MAU), which is attached

er
directly to the 802.3 thick Ethernet cable transmission line.
Thick coax cable (RG-6) requires the following:
• a 50-ohm terminator on each end of the cable;
• a maximum of 3 segments with attached devices (populated segments);

at
• a network board using the external transceiver;
• a maximum of 100 devices on a segment, including repeaters;

y
• a maximum length of 1,640 feet (500 meters) per segment;

op
•
eM
a maximum of 4,920 feet (1500 meters) per segment trunk;
• one ground per segment;
• a maximum of 16 feet (5 meters) between a tap and its device; and

tC
• a minimum of 8 feet (2.5 meters) between taps.
Thick coax cable is inflexible which can make changes, such as adding or mov-
ing workstations and devices, difficult. It also has some distance restrictions.
No
c

Thin Coax Cable

Thin coax cable, or Thin Ethernet, implemented with T-connectors and termina-
en

tors, such as RG-58 and A/U or C/U, have the following specifications:
Do

• a 50-ohm terminator on each end of the cable;

• a maximum length of 1,000 feet (185 meters) per segment;
• a maximum of 30 devices per segment;
•
se

a network board using the internal transceiver;

r

• a maximum of 3 segments with attached devices (populated segments);

•
fe

one ground per segment;

ea

• a minimum of 1.5 feet (.5 meters) between T-connectors;

• a maximum of 1,818 feet (555 meters) per trunk segment; and
• a maximum of 30 connections per segment.
Pl
Re

Lesson 5: The Data Link Layer 115

 Thin Ethernet cable is flexible; however, bending or twisting it can damage or
break it. If it is broken, the whole segment fails. A thin coax implementation can-
not use drop cables, and requires two coax cables to come from a wall in every
Two pieces of thin coax person’s office, which can be a nuisance. Also, the more devices you have, the
cables come into each harder it is to make changes to your network layout. 10BASE2 uses thin Ethernet
person’s office or cubicle;
cable.
one goes to each neighbor’s

ial
office as the coax makes its
way around the office.
Twisted-pair Ethernet Cable
Twisted-pair Ethernet cable has the following specifications:
• a maximum of 1,024 attached workstations;

er
• a maximum of 4 repeaters between communicating workstations; and
• a maximum segment length of 328 feet (100 meters).
Devices that use an unshielded twisted-pair implementation generally connect the
cable to a wall jack. From there, permanent cable drops connect to a punch-down

at
block, which, in turn, connects to a patch panel. Hubs combine and repeat the
signals, and can cascade to facilitate large numbers of connections.

y
op
eM
tC
No
c

Figure 5-8: Twisted-pair Ethernet cable.

UTP configurations employ a hierarchical star topology. They are easily modified
en

and have few limitations.

Do

The 5-4-3 Rule

You can use the 5-4-3 rule when designing your coax or UTP Ethernet
se

configurations. It is a network design guideline that can help you establish an

r

efficient Ethernet network.

Using internetworking
fe

devices, such as routers, you Coaxial implementation: five cable segments per network, connected by up to
ea

can overcome limitations four repeaters, allowing three populated segments.

such as the maximum
network length. UTP implementation: five cable segments per network, connected by up to four
hubs, or multiport repeaters.
Pl
Re

Note: A UTP implementation can be referred to as the 5-4 rule because UTP seg-
ments cannot have attached devices.

The following tables list the IEEE 802.3 specifications:

116
IEEE 802.3
specifications 10Base2 10BaseT3 10Base5
Data rate 10 Mbps 10 Mbps 10 Mbps
Maximum segment length 185 meters 100 meters (distance 500 meters
from hub to
computer)

ial
Minimum distance between .5 meter N/A 2.5 meters
stations
Maximum distance 500 meters N/A 500 meters
between stations
Maximum total network 925 meters 500 meters per 2460 meters

er
length collision domain
Maximum connected 5 (using 4 repeaters) 5 (using 4 hubs in a 5 (using 4 repeaters)
segments with only 3 segments cascading with only 3 segments
allowed to have configuration) allowed to have
computers connected1 computers
connected1

at
Maximum computers per 30 with a further limit 12 per hub with a 100
segment of 1024 per network further limit of 1024
per network

y
Media type Thinnet Twisted pair Thicknet

op
Signaling Baseband
eM
Baseband Baseband
Media Access scheme CSMA/CD CSMA/CD CSMA/CD
Topology Bus Star-wired bus Bus

tC
Connector to network BNC T-adapter RJ-45 DIX or AUI
adapter
Impedance 50 ±2 Ω (Ohms) 85-115 Ω with UTP 50 ±2 Ω
or 135-165 Ω with
STP
No
Terminator resistance 50 Ω (Ohms) 2
50 Ω
c

Footnotes:
en

1: This relationship is often called the ″5-4-3 rule.″

Do

2: Twisted-pair segments are terminated; however, you do not need to worry

about implementing this termination. Hubs provide internal termination for
twisted-pair networks when necessary.
3: The fiber-optic variation of 10BaseT is called 10BaseF, which includes
se
r

three sub-types: 10BaseFL (fiber link), 10BaseFP (fiber passive), and

10BaseFB (fiber backbone).
fe

IEEE 802.3
ea

specifications 100BaseVG
(continued) 100BaseT3 AnyLAN4 1Base5 10Broad36
Data rate 100 Mbps 100 Mbps 1 Mbps 10 Mbps
Pl
Re

Lesson 5: The Data Link Layer 117

 IEEE 802.3
specifications 100BaseVG
(continued) 100BaseT3 AnyLAN4 1Base5 10Broad36
Maximum segment length 2000 meters 100-200 500 meters 3600 meters
(for 100BaseFX) meters for (distance from
less for other twisted pair, hub to

ial
sub-types 2000 meters computer)
for fiber optic
(distance from
hub to
computer)
Maximum distance 205 meters (for N/A N/A 500 meters

er
between stations 100BaseTx and
T4)
Maximum total network N/A 500-1000
length meters for
twisted pair,

at
10,000 (10
km) for fiber
optic

y
Maximum connected N/A 5
segments

op
Media type
eM Twisted pair or Twisted pair Twisted pair Thicknet
fiber optic and fiber optic (broadband)
Signaling Baseband Baseband Baseband Broadband
Media Access scheme CSMA/CD Demand CSMA/CD CSMA/CD

tC
Priority
Topology Star-wired bus Star-wired bus Bus Bus
Connector to network RJ-45 RJ-45 RJ-45 DIX or AUI
adapter
No
Impedance 50 ±2 Ω
c

Terminator resistance 2 2 2
50 Ω
en

Footnotes:
Do

1: This relationship is often called the ″5-4-3 rule.″

2: Twisted-pair segments are terminated; however, you do not need to worry
about implementing this termination. Hubs provide internal termination for
twisted-pair networks when necessary.
se
r

3: The 100BaseT specification has at least three sub-types: 100BaseTx runs

over two-pair Category 5 cabling; 100BaseT4 runs over four-pair Category
fe

3, 4, or 5 cabling; and 100BaseFx runs over fiber-optic cabling.

ea

4: 100BaseVG is a technology developed by Hewlett-Packard. It was stan-

dardized by the IEEE as 802.12. The ″VG″ in the name stands for ″voice
grade.″
Pl
Re

118
TASK 5C-2:
Distinguishing between different Ethernet media
types
Objective: Determine which Ethernet type, thick, thin, or twisted pair, is

ial
appropriate for each of the following circumstances.

1. 900 feet between segments; 2 feet between T-connectors; 3 populated

segments; and 25 connections per segment.

er
2. 300 feet between segments; 350 total connections; star topology; and 2
repeaters between workstations.

at
3. 1,000 feet between segments; 2 populated segments; 100 devices per seg-
ment; and 2 repeaters between workstations.

y
op
4.
eM
Which type of Ethernet is flexible? How flexible is it?

CHECK YOUR SKILLS 5-2 tC Suggested time:

No
2 minutes
c

Using the 5-4-3 rule

Using the 5-4-3 rule, determine the appropriate number.
en
Do

1. How many repeaters can a coaxial implementation have?

2. How many populated segments?

se
r

3. How many cable segments per network?

fe
ea
Pl
Re

Lesson 5: The Data Link Layer 119

 Large networks incur frequent problems with device access and other bottlenecks
because of standard Ethernet’s 10 Mbps limitation on information transfer. New
technologies that are bringing higher CPU performance, high-speed buses, and
high-end applications insist on additional bandwidth. To address this issue, the
Fast Ethernet Alliance has devised IEEE 802.3u Fast Ethernet specifications, such

ial
as physical and logical topologies, Auto Negotiation (AUTONEG), Media Inde-
pendent Interface (MII), and Media Access Control (MAC). These specifications
are summarized in the following table.

Fast Ethernet specification Explanation

er
Topologies Physical—uses a hierarchical star configuration consisting of
individual cables connecting at a common point. Common points
interconnect by using repeaters, hubs, and switches. Logical—
uses a bus configuration.
MII Enables Fast Ethernet to use either Physical Layer Devices
(PLDs), or transceivers (100BASE-TX, 100BASE-T4, or 100BASE-

at
FX). A Fast Ethernet network board can connect to an external
transceiver with UTP cables, creating a flexible environment with
both optical fiber and copper media. Fast Ethernet specifies a

y
maximum copper segment length of 100 meters. However, the
maximum cable length between repeaters is 5 meters. A Fast

op
eM Ethernet network defined by 100BASE-FX and in half-duplex
mode has a maximum segment length of 412 meters without
repeaters—full duplex mode has a maximum length of 2,000
meters.

tC
AUTONEG During a transition to a Fast Ethernet environment, 10/100 Fast
Ethernet adapters can be installed on workstations to establish
(negotiate) a data rate in the mixed environment.
MAC No changes were made to the CSMA/CD protocol.
No
c

Fast Ethernet Physical Specifications

en
Do

Let’s examine some of the physical specifications for various cabling and repeat-
ers used in Fast Ethernet implementations.

100BASE-TX
se
r

The 100BASE-TX specification uses two pairs of Category 5 UTP or Category 1

fe

STP cabling at a 100 Mbps data transmission speed. Each segment can be up to
ea

100 meters long. UTP cables use standard RJ-45 connectors; STP cables use
DB-9 connectors.
Pl
Re

100BASE-T4
The 100BASE-T4 specification uses four pairs of Category 3, 4, or 5 UTP
cabling at a 100 Mbps data transmission speed with standard RJ-45 connectors.
Each segment can be up to 100 meters long.

120
100BASE-FX
The 100BASE-FX specification uses two-strand 62.5/125 micron multi- or single-
mode fiber media. Half-duplex, multi-mode fiber media has a maximum segment
length of 412 meters. Full-duplex, single-mode fiber media has a maximum seg-
ment length of 10,000 meters. 100BASE-FX includes the use of Media Interface
Connectors (MIC) or Subscriber Connectors (SC) specified by ANSI FDDI.

ial
Repeaters
There are two types of Fast Ethernet repeaters, classified by performance and fea-
tures: Class I and Class II. Both are 100 Mbps repeaters. One Class I repeater or

er
two Class II repeaters can be used in a single collision domain.
IEEE 802.3u repeaters sometime have to translate frequencies when interconnect-
ing dissimilar 100BASE segments. For example, 100BASE-TX and 100BASE-FX
both use a 41.6 MHz signal frequency, but 100BASE-T4 uses a 25 MHz signal

at
frequency. The repeater must translate the frequencies to connect the segments
that use different frequencies. The maximum number of repeaters per collision
domain is two with a maximum inter-repeater link cable length of 5 meters.

y
Note: A repeater that translates frequencies brings more latency to a network than

op
eM
does a single frequency repeater. Latency is the amount of time it takes a packet to
travel from sender to receiver.

TASK 5C-3:
Discussing Fast Ethernet
tC
No
latency:
c

1. Why was Fast Ethernet developed? The amount of time it takes a

packet to travel from sender
to receiver.
en
Do

2. What type of media does each of the Fast Ethernet physical specifica-
tions require?
se
r
fe

3. True or false: A repeater that translates frequencies increases network

ea

latency.
Pl
Re

802.5 and Token Ring

IBM Token Ring is essentially a subset of IEEE 802.5. Because IBM has pro-
vided attractive mainframe-to-LAN connectivity solutions for Token Ring
networks, Token Ring has become a popular lower-layer standard.

Lesson 5: The Data Link Layer 121

 Like 802.3, 802.5 provides MAC sublayer and Physical layer functions by using
802.2’s link control functions

ial
er
at
y
op
eM
Figure 5-9: An IEEE 802.5 token-passing ring. Station A transmits a frame to Station C.

802.5 Media Access Control Sublayer tC

No
IEEE 802.5 uses token passing to control access to the medium. IEEE 802.5
includes a priority scheme that enables specific nodes to control the token (and
c

access to the medium) longer than other nodes.

en
Do

Active Monitor
IEEE 802.5 requires that there be a token somewhere on the network at all times
and that stations relinquish the token when they have finished using it. To enforce
these requirements, one of the stations on the network functions as an active
monitor; any station can be the active monitor. If no active monitor exists, a sta-
se
r

tion is automatically designated.

Suppose that the station holding the token fails (loses power, for example) before
fe

it can transmit a data frame or pass the token on to the next station. Stations can-
ea

not transmit until they have the token, so no station would be able to transmit.
The same situation might occur if the token were damaged in transmission.
The active monitor constantly examines the network. If a certain interval passes
Pl
Re

and the active monitor does not detect any frames (the token or a data frame) on
the network, then the active monitor determines that the token has been lost and
generates a new token.

122
Persistently Busy Token
After a station transmits a data frame, that frame travels around the ring, eventu-
ally returning to the station that transmitted it. At that point, the station
relinquishes the token. While the token is in the possession of a station, it is said
to be ″busy.″

ial
Consider what would happen if the station holding the token failed just after it
transmitted a frame. The station would not be able to release the token to the
next station, and the data frame would travel around the network endlessly unless
it was stopped. In this case, the active monitor would mark the data frame by
setting its monitor bit (in the frame’s Access Control field). If the marked frame
were to return to the active monitor, the active monitor would assume that there

er
was a problem. It would remove the data frame from the network and pass a new
token to the next station.
Stations other than the active monitor function as passive monitors; they monitor
the activities of the active monitor. If the active monitor fails, one of the passive
monitors takes its place.

at
Priority Control

y
An optional feature called priority control enables stations to have channel-access

op
eM
priority over other stations. This is implemented through three priority bits and
three reservation bits in the Access Control field. Stations use the reservation bits
to reserve tokens. The priority of the frame being transmitted is held in the prior-
ity bits.

TASK 5C-4: tC
No
c

Discussing 802.5’s MAC sublayer

1. What two things does IEEE 802.5 require?
en
Do

2. What happens if the token is not passed to the next node?

se
r

3. What is the active monitor’s function?

fe
ea

4. True or false: It is not possible to prioritize node access to the network.

Pl
Re

Lesson 5: The Data Link Layer 123

 802.5 and the Physical Layer
Although IEEE 802.5 does not specify a particular topology, IBM’s Token Ring
has established the ring topology as the most popular. Token Ring stations con-
nect to a Multi-Station Access Unit (MSAU). Relay switches inside the MSAU
bypass ports to which no station or a defective station is connected to provide a
continuous ring. Because the ring path is hidden inside the MSAU, Token Ring

ial
networks look like star topologies. Because of this, Token Ring networks are
sometimes described as a ″star-wired ring.″

er
at
y
op
eM
tC
Figure 5-10: Because the ring ″path″ is hidden inside the MSAU, Token Ring’s topology
appears more like a star than a ring.
No
Adapter cables are used to connect MSAUs in the main ring path. By using patch
c

cables to connect MSAU’s ″ring in″ and ″ring out″ ports, you can link several
MSAUs to provide connections for more stations than a single MSAU permits.
en
Do
se
r
fe
ea
Pl
Re

Figure 5-11: Two MSAUs connected with a patch cable to enable more nodes to connect.

124
802.5 Token Ring specifications
Lobe length (distance from node to 330 ft (100 m) for Type 1 or 2 cable; 220 ft (66m) for Type 6
MSAU) or 9; 150 ft (45 m) for Type 3 UTP
Distance between nodes 8 ft (2.5 m) minimum
Ring length (distance between MSAUs 660 ft (200 m) for Type 3 cable; 400 ft (45m) for Type 6;6
on main path) mile (1 km) for optical fiber

ial
Maximum number of nodes 250 (IEEE 802.5); 260 (IBM STP); 72 (IBM UTP)
Maximum number of segments 3
between repeaters
Maximum number of MSAUs 33
Type of media None specified

er
Media access method Token passing
Data rate (in megabits/second) 4 or 16 (16 requires Type 4 or better cable)

In addition to the specifications listed above, Token Ring requires that each seg-

at
ment is terminated at both ends and grounded at one.

Be sure not to confuse 802.

5, the Token Ring

y
Beacon Frames specification, with 802.4, the
Token Bus specification.

op
eM
IEEE 802.5 includes a feature that helps the network to automatically recover
from errors, such as breaks in the network cabling. Stations detecting a serious
network error continuously transmit beacon frames around the ring until they
receive one from a station preceding them. A beacon frame enables other stations

tC
to isolate the area in which the problem occurred (the failure domain). Once a
station receives a beacon frame from a station preceding it, it stops sending bea-
con frames. When only one station is left sending beacon frames, the problem is
isolated to the station preceding the station sending frames. The station still send-
No
ing beacon frames sends a beacon MAC frame containing its addresses, its
c

nearest active upstream neighbor (NAUN), and all stations in between that could
have failed.
The network can be automatically reconfigured to remove the failure domain
en

(which includes the station that transmitted the beacon frame, the station preced-
Do

ing it on the network, and the medium between them) from the network.
The following table lists the IEEE 802.5 and Token Ring specifications.

IEEE 802.5 and Token

se

Ring specifications
r

IEEE 802.5 Token Ring

fe

Data rate 1 or 4 Mbps 4 or 16 Mbps

ea

Minimum distance between not specified 2.5 meters

stations and MSAU
Maximum distance between not specified 101 meters (with Type 1 cables),
stations and MSAU (lobe cable 100 meters (with STP), or 46
Pl
Re

distance max.) meters (with UTP)

Maximum distance between not specified 152 meters (500 ft)
MSAUs
Maximum ring length not specified 160 meters (16 Mbps over Type
1 cables) to 360 meters (4 Mbps
over Type 1 cables)

Lesson 5: The Data Link Layer 125

 IEEE 802.5 and Token
Ring specifications
IEEE 802.5 Token Ring
Maximum connected segments not specified 33 MSAUs (9 over Type 3 cable)
Maximum number of stations 250 260 with STP or 72 with UTP

ial
Media type UTP or STP UTP or STP
Signaling Baseband Baseband
Media Access scheme Token passing Token passing
Topology None specified, though Star-wired ring
typically a star-wired ring
Impedance 100-120 Ω UTP 150 Ω STP 100-120 Ω UTP 150 Ω STP

er
Terminator resistance N/A N/A
Connector to network adapter not specified Media Interface Cable (MIC) (an
IBM proprietary ″genderless″
connector), RJ-45, or RJ-11

at
y
TASK 5C-5:

op
eM
Discussing 802.5’s Physical layer
1. What is the most widely used physical topology for IEEE 802.5?

2. tC
Why do 802.5 token ring topologies look like star topologies?
No
c

3. What is the purpose of a beacon frame?

en
Do
se

Summary
r

In this lesson, you’ve examined the Data Link layer and the two subdivi-
fe

sions of the layer, MAC and LLC. You also examined IEE 802 standards for
ea

802.2, 802.3, and 802.5.

Pl
Re

126
LESSON 5 REVIEW
5A List some of the functions of the Data Link layer.

ial
5B What is the purpose of the MAC layer?

Describe how communication is handled with the upper layers.

er
Describe LLC’s connection services.

at
y
op
eM
5C Provide a general description of IEEE 802.3, and describe the differ-
ences between IEEE 802.3 and Ethernet.
tC
No
c
en

List the specifications of IEEE 802.5, and describe the difference

Do

between IEEE 802.5 and Token Ring.

se
r
fe
ea
Pl
Re

Lesson 5: The Data Link Layer 127

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Data Transmission LESSON
6

ial
Overview Data Files:
In this lesson, you’ll examine the hardware used in data transmission (the Network+ Certifiication
Physical layer of the OSI Model). Then you’ll examine how that data is put CD-ROM

er
on the wire (the Data Link layer of the OSI Model).
Lesson Time:
1 hour, 30 minutes

Objectives

at
To examine how data is placed on the data-transmission media, a function pro-
vided by the Physical and Data Link layers of the OSI model, you will:

y
6A Examine network connection devices.

op
eM
The purpose of network connection devices and issues associated with
them is covered in this topic.

6B Identify the role of network adapters in data transmissions.

tC
You’ll examine several bus types and the associated network adapters for
each. You’ll also look at what the network adapter is responsible for in
data transmissions. Finally, you’ll examine how to install, configure, and
troubleshoot network adapters.
No
c

6C Examine media-access control methods.

In this topic, you’ll examine contention, polling, token passing, and
demand priority media access control methods.
en
Do

6D Identify the characteristics of common network topologies.

This topic revisits network topologies and examines the advantages and
disadvantages of bus, ring, star, mesh, and hybrid topologies.

6E Describe different types of bridges and how they route data through a
se
r

network.
In this topic, you’ll examine transparent, source-routing, and source-
fe

routing transparent bridges. You’ll learn about the function provided by

ea

bridges and how they affect network traffic.

Pl
Re

Lesson 6: Data Transmission 129

 TOPIC 6A
Introduction to Network Connection Devices
Network connection devices connect the various segments of network media

ial
(cables, optical fiber, radio waves, and so forth) that make up a network or
internetwork. A segment is one section of network transmission media that is
Examples of concentrators assigned a specific network address; it provides servers and clients on the net-
include hubs, MSAUs, and work with access to network resources.
switches. Each of these
devices enables several When connecting several media segments to form a large network, you can opt to
devices to connect to it, thus use one or more network connection devices. A network connection device deter-

er
concentrating access of mines whether connected segments are on the same or different networks by
multiple devices to a single
device.
assigning them the same or different network addresses.
Several types of network connection devices exist, including repeaters, bridges,
switches, and routers . These are general names; when attempting to purchase one

at
of these devices, you might find it under another name. For example, repeaters
might also be called hubs or concentrators . There are also some less common
network connection devices, such as brouters (a bridge/router hybrid) and gate-

y
ways . Each type of device solves a specific set of network connection problems.
concentrator:

op
A general term for hubs,
MSAUs, and switches, all of
eM
which concentrate access of
multiple devices to a single
device.

tC
No
c
en
Do
se
r
fe
ea

Figure 6-1: Network connection devices manage network traffic problems.

Pl
Re

Connection Device Hardware

For devices to communicate and be networked, they must be attached to trans-
mission media. Connection hardware, such as network interface boards,
transmission media connectors, and modems, join computer devices to media
cable.

130
 ial
er
at
y
op
Figure 6-2: Network interface boards.
eM
tC
Network Interface Board
A network interface board attaches a device, such as a computer, to a transmis-
sion medium. It provides the necessary circuitry to make the connection
No
(physically and logically). The majority of the board is installed in the computer;
c

however, one end of it remains outside the computer so that the transmission
media’s connector can be attached to it, as shown in Figure 6-2. Every computer
on a network must have its own network interface board.
en

Types of network interface boards include:

Do

• Network interface cards (NIC)

• Certain external devices (transceivers)
• Software with a generic hardware port
se
r

Transceiver
fe
ea

The transceiver is the basic component of a NIC, in that this is what enables the
NIC to send and receive data.
Used by most network interface boards, transceivers send and receive EM or
electric signals over the transmission media. They are physically connected to the
Pl
Re

transmission media by a connector. Transceivers for wireless media are simply

sending and receiving devices for which no mechanical connectors are necessary.
A transceiver is often a separate box connected to a ″pigtail″ cable.

Lesson 6: Data Transmission 131

 If you need to connect dissimilar media without changing the access method, you
can use a transceiver. For example, if you need to connect a 10BaseT segment to
a 10Base5 backbone, you could use a transceiver since they both use Ethernet.
Transceivers are most often found in 10Base5 environments to connect worksta-
tions to the 10Base5 backbone.

ial
Network Interface Card
Also known as network adapters, network interface cards (NICs) are circuit
boards that provide the means to change a computer’s electric signals to those
used by the transmission medium. Sometimes a computer does not provide a net-
work port. A NIC is used to add this capability to a computer. NICs can contain

er
one or more types of media connectors, as shown in Figure 6-3.
Typically, a NIC uses an internal transceiver built into its circuitry. However,
external transceivers can be attached to the NIC’s cable or media connector, if
necessary.

at
These days, ″cards″ generally refers to PC or PCMCIA cards, like those used in
notebook PCs, which enable a computer to make a network connection. NIC is
an old expression that is synonymous with network interface board.

y
op
eM
Transmission Media Adapter
Sometimes a network interface board uses a connector different from those

tC
already in use by the transmission medium. When this happens, the signals from
one connector can’t communicate with those from the other connector. To prevent
this, a transmission media adapter is used. A transmission media adapter receives
signals from one of the connectors and converts them to a format interpretable by
the other connector.
No
transmission media
c

adapter:
Receives signals from one of
the connectors and converts
en

them to a form interpretable

by the other connector.
Do
se
r
fe
ea
Pl
Re

132
 ial
er
at
y
op
eM
tC
Figure 6-3: Various transmission media connectors.

Transmission Media Connector

No
c

A transmission media connector joins a network device to a media cable. Each

media type has a compatible transmission connector (see Figure 6-3). Some
media have more than one from which you can choose. Media connectors attach
en

to media cable through the network interface board. transmission media

Do

connector:
Joins a network device to a
media cable.
Modem
Computers use electric pulses to send data signals, but some media, such as
se

microwave, as well as public phone networks, use EM waves. This presents a

r

problem when trying to communicate. To enable communication between devices

that use electric signals and EM waves, a modem is used. A modem is a
fe

MOdulator/DEModulator that changes electric (digital) signals from a computer

ea

to analog signals for a microwave transceiver or phone line.

modem:
A modem can be used to do the following: MOdulator/DEModulator that
• Boost signals that are too weak to travel from the transceiver to the destina- changes electric (digital)
Pl
Re

tion without losing significant amounts of data along the way. signals from a computer to
analog signals for a
• Use alternate EM frequency bands if multiple communications are taking microwave transceiver or
place on the same transmission medium. phone line.
• Replace a NIC when connecting to a network (when using a phone line to
connect to the network).

Lesson 6: Data Transmission 133

 When remote access is set up on a network, a modem must be configured on the
network that gives users access to network resources. However, if two devices are
within tens of meters of each other, a cable wired in a null modem, or ″dumb″,
configuration can be used in place of a modem.

ial
er
at
y
op
eM
Figure 6-4: Network connection devices and the OSI model.
As you can see in Figure 6-4, each kind of network connection device operates

tC
within specific layers of the OSI reference model. The capabilities and limitations
of each device are directly related to the OSI layers in which the device operates.
Some of these devices are network devices; they connect segments of networks to
each other and are only aware of local network addresses. Devices that operate
above the Data Link layer are thought of as internetwork devices. They connect
No
networks to each other and are aware of addresses outside of the local area
c

network.
In the sections that follow, you will learn the capabilities and limitations of
en

internetworking devices, and you will learn the roles that each device plays in an
Do

internetwork. In practice, the distinctions between these devices are sometimes

unclear. Manufacturers often combine favorable attributes of two or more types of
devices to satisfy the needs of their customers.
se
r
fe
ea
Pl
Re

134
 ial
er
at
Figure 6-5: Repeaters connect two network segments.

y
op
Repeaters
eM
The main purpose of repeaters is to overcome the affects of attenuation. Repeat-

tC
ers work in the Physical layer of the OSI model to provide fast signal
regeneration.
Because of attenuation (loss of signal strength over distance), each type of trans-
mission media used in networks has a maximum suggested length. To extend the
No
network beyond this distance, you can use repeaters. A repeater is typically used
c

as a connection device between two segments on the same network. It can also
be used, however, as a transmission media adapter that connects different media
types. repeater:
A device that connects two
en

network segments to make

Do

them work as one. Repeaters

can extend the length of a
network beyond the physical
limitations of a single cable.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 135

 ial
er
at
y
Figure 6-6: A repeater regenerating a weak signal.

op
eM
Amplifiers
In digital computer networks, repeaters aren’t simply amplifiers; they’re signal

tC
regenerators. If repeaters only amplified a signal, they would also amplify any
noise on the data-transmission medium. Instead, repeaters deal with attenuation
This would be like turning by receiving and evaluating the data (much like any network computer) then
the volume up while immediately re-transmitting a clean copy of the signal on to the next segment of
No
listening to a distant radio
the network cable.
station. The signal gets
c

louder, but so does the

static.
Signal-regenerating Repeaters
en
Do

This type of repeater boosts a signal and produces less attenuation by receiving
the data and removing it from the transmission signal. It then puts the signal back
together and resends it on the next network cable segment. The signal is an exact
replica of the original, including its original strength.
Regenerating a signal takes very little time. However, LANs don’t function well
se
r

if a significant amount of time elapses between signal transmission and reception.

This period of time, called a propagation time delay , shouldn’t exceed a specific
fe

limit. Introducing several repeaters into a network can cause an excessive propa-
ea

gation time delay. For this and other reasons, there is a limit on the number of
repeaters that a network can use. This number is called the repeater count . The
repeater count depends on the type of data-transmission medium and other related
factors. Although the repeater count you can use is limited, you can use repeaters
Pl
Re

to extend the span of a LAN well beyond the normal limitations of the data-
transmission medium. Network segments connected by a repeater must be
identical (for example Ethernet to Ethernet or Token Ring to Token Ring).
Devices that perform the functions of a repeater include Ethernet multiport
repeaters, active hubs, and concentrators.

136
One common type of repeater amplifies the signal and is often referred to as an
amplifier ; another type regenerates the signal and is often referred to as a signal
regenerating repeater .

ial
er
at
y
op
eM
Figure 6-7: Bridges can filter inter-segment traffic.

Bridges
tC
A bridge , like a repeater, extends a network’s transmission distance. However, a
No
bridge is used to connect segments on different networks. It uses a recipient
node’s address to determine which signals should be passed to another network
c

segment (see Figure 6-7).

For example, if the bridge shown in Figure 6-7 receives a data unit from Segment
en

1, it examines the destination node address for that data unit. If the data unit is
Do

destined for a node on Segment 1, the bridge ignores the data unit; the data unit
has already found its way to the correct network segment. If the data unit is des-
tined for a node on another segment, then the bridge forwards the data unit to the
next segment.
In this scheme, traffic local to a particular segment is confined only to that seg-
se
r

ment; however, internetwork traffic is allowed to pass through the bridge. You can
use bridges when you need to restrict traffic across network segments.
fe
ea

Hubs
Depending on how your network is configured, you might need a single point
Pl
Re

where media segments can connect. This single point is called a hub . A hub acts
as a go-between for devices connected to a network; it receives transmission sig-
nals from one device and sends them out to other connected devices. It has
multiple ports to which network devices attach.
There are different kinds of hubs, including active and passive hubs, multiport
repeaters, and switches.

Lesson 6: Data Transmission 137

 Active Hub
Like a repeater, an active hub can regenerate or amplify a signal. Because of this,
and because it connects multiple media segments, the active hub extends cable
length and thereby increases transmission length. Computers attached to an active
hub continue to receive transmissions from all other network computers. In other
words, the hub does not perform any filtering of packets; it broadcasts everything

ial
over every segment.

Passive Hub
A passive hub resembles an active hub only in that it connects multiple media

er
segments together. Aside from that, they differ greatly:
• A passive hub is not a repeater; it does not regenerate or amplify a signal.
• Restrictions placed on segments attached to a passive hub are different from
those placed on segments attached to an active hub.

at
• Additional restrictions can be placed on segments attached to a passive hub.
For example, prohibiting two passive hubs from being consecutively
connected.

y
op
eM
Multiport Repeater
A multiport repeater is similar to an active hub; it receives signals and regener-

tC
ates them. However, a multiport repeater transmits the signals not only to the
intended receiver, but also to every port, regardless of address.
No
c
en
Do
se
r
fe
ea
Pl
Re

138
 ial
er
at
y
op
eM
Figure 6-8: Switches control which segments are connected to each other.
tC
No
Switch
c

Contrary to a multiport repeater, a switch receives a transmission and retransmits

it using only that port through which the intended receiver can be reached. This
en

is very similar to a bridge.

Do

Switches can make efficient use of bandwidth; all of a network’s media segments
can be permanently connected. Each segment is used only for those transmissions
targeted at a computer on that particular segment.
se
r

Multiplexers
fe

Sometimes bandwidth is not used efficiently because a signal is not large enough
ea

to use all the space provided by the transmission media. To avoid this, you can
install a multiplexer, which runs multiple signals on a single transmission
segment.
Pl
Re

Lesson 6: Data Transmission 139

 Multiplexing
In many cases, data-transmission facilities can be compared to postal services.
For example, consider how expensive it would be for people who have never
mailed a letter or packet to hire their own private postal workers, purchase their
own delivery trucks, airplanes, and so forth. Furthermore, consider how those
facilities would be wasted. With the small volume of mail that one customer

ial
could generate, those facilities would be idle most of the time.
As ridiculous as this scenario seems, this type of situation could easily happen
with data-transmission facilities. Often, a main communication line provides a
high-bandwidth (high-volume) connection between two points. Devices on one
end of the line need to communicate with devices on the other end. None of

er
those devices is likely to need the entire bandwidth of the communication line. It
would be expensive and wasteful to have a dedicated line for each set of devices
that needed to communicate.
Multiplexers enable several devices to simultaneously share a high-bandwidth
line. They do this by combining several low-speed channels at one end into a

at
single high-bandwidth channel (multiplexing or muxing), and then splitting (de-
multiplexing or demuxing) the channel at the other end back into low-bandwidth
channels. A multiplexer at either end of the high-bandwidth line performs the

y
multiplexing and de-multiplexing functions. A multiplexer is often referred to as a
mux .

op
eM
To enable devices to share a high-bandwidth line, a multiplexer employs one of
the following schemes:
• Frequency-Division Multiplexing (FDM) —divides a broadband channel’s

tC
wide band of frequencies into several smaller frequency bands of
subchannels.
• Time-Division Multiplexing (TDM) —divides the band on a timesharing
basis.
No
• Statistical time-division multiplexing (STDM or statmux) —designates a time
c

slot for a particular network device according to when the signal accesses
the channel.
en
Do

TASK 6A-1:
Identifying network connection devices
se
r

Objective: Figure out which network connection device is described.

fe

1. Converts digital signals to analog and vice versa.

ea

2. Runs several signals on one media segment.

Pl
Re

3. Connects a device to a media segment.

140
4. Receives signals from a device and transmits them to other devices.

5. Connects segments from different networks.

ial
6. Connects segments from the same network and boosts data signals.

er
CHECK YOUR SKILLS 6-1 Suggested time:
5 minutes
Discussing network connection devices

at
Provide a definition for each of the following devices.

1. What is a media segment?

y
op
eM
2. How does a passive hub differ from an active hub?

3. How does a switch differ from a multiport repeater? tC

No
c
en

4. What is a NIC and why is it used?

Do
se
r

APPLY YOUR KNOWLEDGE 6-1

fe

Suggested time:
ea

5 minutes
Applying networking devices
Decide which networking device is appropriate for each of the following
scenarios. Explain why you chose that particular networking device.
Pl
Re

1. McCloskey Associates needs to provide occasional communication between

segments. Although network traffic is high, most communication is among
nodes on the same segment.

Lesson 6: Data Transmission 141

 2. Fenster Co., whose offices occupy two floors of a small building, needs to
extend the length of its Ethernet cable system to connect two offices. There
are no standard network traffic patterns.

ial
3. Condello Concrete Company has an old Ethernet network. They would like
to use their existing network bandwidth better and make it easier to modify
traffic patterns as needed.

er
4. KK Software Corp. has a room in which new software is tested. The eight
computers in the room run their own network and need a device through
which they can connect to each other.

at
TOPIC 6B

y
op
eM
Network Adapters
Network adapters prepare data for the data-transmission media, send data to and

tC
receive data from the network, and translate data into bytes the computer can
understand. Network adapters can also be called network interface cards (NICs),
LAN cards, and interface cards .
No
c
en
Do
se
r

Figure 6-9: A network adapter.

fe
ea

The network adapter fits into a slot in the computer called a bus , and is attached
to the network cable via a connector. The purpose of the bus is to shuttle data
back and forth from a network adapter to memory and on to the central process-
ing unit (CPU). A bus can be defined as the internal pathway that connects the
Pl
Re

microprocessor to random access memory (RAM). Buses are measured by the

number of bits that can be transferred at one time. The larger the number of bits

142
that are transferred at one time, the faster the bus is. Older buses transmitted 8
bits at a time and were known as 8-bit buses. Newer buses are 16-bit and 32-bit
buses. The type of bus in the computer determines which type of network adapter
must be used to connect to the network. There are several types of buses, as
explained in the following table.

Bus abbreviation Bus architecture Description

ial
It is inappropriate to
- PC Bus The original 8-bit bus of the IBM PC and abbreviate MicroChannel
XT computers. Architecture as MCA. MCA
ISA Industry Standard 16-bit system developed in 1983; a is a trademark of Universal
Architecture majority of computers use this design; less Studios, who legally
expensive than EISA. challenged IBM’s use of the
acronym.

er
EISA Extended Industry Standard 32-bit design; compatible with ISA
Architecture standard.
- MicroChannel Architecture IBM proprietary 32-bit design; a very fast
architecture.
NuBus Macintosh NuBus Apple’s 32-bit design; used only on

at
Macintoshes.
VESA Video Electronics Standards 32-bit design; developed to accelerate
Association video displays.

y
PCI Peripheral Component A more comprehensive 32-bit design; the
Interconnect first to offer Plug And Play setup.

op
PC Card PC Card
eM
Integrated circuits that enable a notebook
computer with one port to interface with
several different devices; about the size of
a credit card; fits into a slot in a mobile

tC
computer; has the ability to ″hot swap,″ or
change the card without restarting the
computer. Originally, this bus was known
as the Personal Computer Memory Card
No
International Association (PCMCIA) bus.
c

However, that name was dropped in favor

of the much simpler PC Card.
en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 143

 ial
er
at
y
op
eM
Figure 6-10: Examples of buses.

tC
No
c

Figure 6-11: A PC Card network adapter.

en
Do

Data moves through a computer’s bus 8-bits, 16-bits, or 32-bits at a time across
separate and parallel conductors. Such a transmission method is called parallel
because you could think of the bits as ″moving side-by-side.″ Network cabling,
however, transmits data in serial fashion, one bit after the other down a single
conductor. It is the job of the transceiver (transmitter/receiver) component of the
se

network adapter to translate the parallel data stream of the bus to the serial data
r

stream the network cable can use. Some network cards require an external trans-
ceiver; others come with an on-board transceiver. Typically, adapters that have an
fe

on-board transceiver also support an external transceiver for backward

ea

compatibility.
Pl
Re

144
 ial
er
Figure 6-12: A network adapter with an on-board transceiver.

Wireless Network Adapters

at
Network adapters can also support wireless networks, which means the network
adapter doesn’t need to be physically connected to a network cable as its data-

y
transmission medium. Wireless network adapters for radio-frequency
transmissions often come with an indoor omnidirectional antenna to function as

op
the data-transmission medium.
eM
tC
Role of the Network Adapter
Network adapters are responsible not only for translating data into a form the
network cable can transmit, but also for broadcasting their address to the network. The MAC in MAC address is
Each network adapter has a unique hardware address, often called a MAC (Media named after the Media
No
Access Control) address , which is used to distinguish it from the rest of the Access Control sub-layer of
c

adapters on the network. Most network adapters have their hardware addresses the Data Link OSI Model
assigned to them at the time of manufacture. layer.
en

An IEEE committee determines MAC addresses. Blocks of addresses are assigned

Do

to each manufacturer. Manufacturers then assign addresses to each card from their
assigned pool of addresses. Each card thus has a unique address. A MAC address
is a 6-byte hexadecimal number (at least for Ethernet cards, other technologies
might vary); for example, 00-A0-8F-CA-3C-25.
The network adapter also communicates with the receiving network adapter to
se
r

determine the following information:

• The maximum size of the groups of data to be sent.
fe

• The intervals between groups of data.

ea

• The amount of data that will be sent before the receiving card issues a
confirmation.
• The wait time for confirmation.
Pl
Re

• The data capacity of each card.

Lesson 6: Data Transmission 145

 • The data-transmission speed.
Finally, network adapters assemble data into frames. Frames are the smallest unit
frames: of data sent over the network. Assembling frames is a function of the Data Link
The smallest unit of data layer of the OSI Model. A typical frame has the following components:
sent over the network.
Component Description

ial
Destination ID The MAC address of the device that is to receive this frame.
Sender’s ID The MAC address of the device that sent this frame.
Control Information about the frame, such as its type and segmentation information
(how the data was divided into frames).
Data The data being sent, or at least part of it.

er
Cyclic Redundancy Check An error detection code typically computed by some sort of mathematical
(CRC) operation on the data (such as adding it up). The CRC is re-calculated at
the receiving station to determine if the frame was corrupted en route. If
so, a re-transmission of the data is requested.

at
Performance

y
Although each network adapter is set to meet certain standards, some network

op
eM
adapters provide features that increase the data-transmission speed. The following
table shows the features that can improve network performance.

Feature Description

tC
Shared adapter memory RAM on the network adapter shared by the adapter and the
computer.
Shared system memory A section of computer memory used by the network adapter to
process data.
No
Bus mastering Temporary control of the bus by the network adapter; data
c

bypasses the CPU and moves directly into system memory.

RAM buffering A buffer created by RAM chips on the network adapter to hold
the data overflow. Data overflow occurs when the network
en

adapter receives more data than it can transmit to the computer

Do

across the bus at one time.

Onboard microprocessor A microprocessor contained on the network adapter that helps
process data without relying on the CPU.
se
r

Network Adapter Configuration

fe
ea

Network adapters, as well as other adapters, hardware, and applications, must

communicate with the CPU in an orderly and established manner. To this end,
network adapters have configuration options that must be set for the network
adapter to function properly. Four of the most common configuration options are
Pl
Re

interrupt (IRQ), base I/O port address, base memory address, and transceiver (for
Ethernet network adapters).

146
Interrupts
Interrupts signal the system processor that the adapter card, system hardware, or
application software needs attention. When signaled, the processor puts its current
task on hold and responds, or services, the interrupt. The Intel family of computer
chips supports hardware, exception, and software interrupts.

ial
Interrupt Description
Hardware Interrupts that come from system hardware, such as the keyboard or the system
clock, or from adapter cards. One such signal, interrupt number 25, causes the
processor to reset and the system to reboot. This is what happens when you
press [Ctrl][Alt][Del].

er
Exception Interrupts used mainly by the processor itself to handle error conditions. An
exception interrupt will be issued, for example, when a process requests the
processor to divide a number by zero.
Software Interrupts sent by an application running on the computer. Software interrupts
generally trigger one of the built-in BIOS routines. These built-in routines can
be replaced by routines available from third-party vendors. The substitute

at
routines would then be called in case of a software interrupt.

y
Adapter cards have a unique hardware interrupt number to distinguish them from
the other adapters in the system. The term IRQ , or interrupt request line , can be

op
eM
used to describe the interrupt number to be used by an adapter. Should two or
more adapters share an interrupt number, the processor will be unable to deter-
mine which adapter sent the interrupt. In such a situation, the processor might
attempt to service either or neither of the adapters, usually resulting in erratic

tC
behavior. (Some new buses support shared interrupts; the adapters used in such
configurations must support the same technique for sharing interrupts.)
In most cases, IRQ3 or IRQ5 can be used for a network adapter. IRQ5 is the
default setting on most computer systems. On Microsoft systems, you can use
No
Microsoft Diagnostic (MSD.EXE or WINMSD.EXE) to determine which IRQs
c

are currently being used.

IRQ Common use in an 80286 and higher computer

en
Do

2 EGA/VGA video adapter

3 COM2, COM4, bus mouse; because these devices aren’t commonly used, this IRQ is
generally available.
4 COM1, COM3
5 LPT2; because this device isn’t commonly used, this IRQ is generally available.
se
r

6 Floppy disk controller

7 LPT1
fe

8 Real-time clock (RTC)

ea

9 Cascade; makes IRQs higher than this one available.

10 Available
11 Available
Pl
Re

12 PS/2 port mouse

13 Math co-processor
14 Hard disk controller
15 Available

Lesson 6: Data Transmission 147

 Input/Output (I/O) Addresses
Once the adapter has signaled the processor that it needs attention (with an inter-
rupt), it then needs a way to communicate with the CPU. This is accomplished
with an I/O address. Several terms are used to describe this feature, such as base
I/O address, I/O address, or I/O port. An I/O address marks the beginning of a
range of memory, usually in the lowest portions of memory, that is used for com-

ial
munication between the CPU and the adapter. Each adapter must have its own
unique, non-overlapping I/O address space. I/O addresses are generally ranges of
memory from 4 to 32 bytes of contiguous memory space. These addresses usually
fall between 0x200 and 0x3FF (hexadecimal). Your computer documentation
should include a list of addresses already in use.

er
Port Typical use Port Typical use
200-20F Game port 300-30F Network adapter
210-21F 310-31F Network adapter
220-22F 320-32F Hard disk controller (PS2 model 30)

at
230-23F Bus mouse 330-33F
240-24F 340-34F
250-25F 350-35F

y
260-26F 360-36F

op
270-27F
eM LPT3 370-37F LPT2
280-28F 380-38F
290-29F 390-39F
2A0-2AF 3A0-3AF

tC
2B0-2BF 3B0-3BF LPT1
2C0-2CF 3C0-3CF EGA/VGA
2D0-2DF 3D0-3DF CGA/MCGA, EGA/VGA color video
No
modes
c

2E0-2EF 3E0-3EF
2F0-2FF COM2 3F0-3FF Floppy disk controller and COM1
en
Do

Base Memory Addresses

The base memory address specifies a location in RAM used to provide a buffer
space (also called a control program execution space) on the adapter. For network
se

adapters, most often the base memory address is used for storing incoming and
r

outgoing data. Memory addresses for these on-board memories are usually
mapped to the range between 640 kilobytes (KB) and 1 megabyte (MB), the
fe

upper memory block region of the CPU. Adapters must have unique base
ea

memory addresses, and must operate in the same mode—either 8-bit or 16-bit.
For example, problems can arise when you use an 8-bit network card with a
16-bit video card. In such a situation, the computer might not be able to boot or
Pl

operate properly due to the mode differences. To correct the problem, you could
Re

either set the video card to operate in 8-bit mode or upgrade the network adapter
to a 16-bit card.

148
Direct Memory Access
Direct memory access (DMA) is a technique used for speeding data transfers
within the computer. In a non-DMA situation, the CPU is involved with transfer-
ring every byte of data between adapters and system memory. While dealing with
these data transfers, the CPU cannot perform other tasks. The DMA controller
works with the CPU and other components in the system to speed data transfer.

ial
The controller relieves the processor from this duty and handles all of the related
issues. The DMA controller also communicates with the adapters to facilitate
transferring data from them directly to system memory. Each adapter must use a
unique DMA channel to communicate with the DMA controller. Failure to do so
will prevent proper operation of the adapters. An 8-bit adapter can use DMA
channels 1, 2, or 3, but channel 2 is generally reserved for the floppy-disk drive

er
controller. A 16-bit adapter can use DMA channels 4 through 7, but channel 4 is
rarely, if ever, used and should be considered reserved.

Configuring Adapters

at
Configuration information can be set by using software, switches, or jumpers
—small connectors that tie two pins together to determine which circuits an

y
adapter will use. On older adapter cards, you physically alter circuitry paths on
the card by selectively opening or closing switches. Most network adapters no

op
eM
longer have to be physically altered to set the switches. Software programs now
allow the network adapter switches to be set through user input in the program.

tC
No
c
en
Do

Figure 6-13: Examples of jumpers and switches for an ISA network adapter.
Each brand of adapter (and in some cases each model from the same manufac-
turer) uses different software for configuration. Generally, these applications use
se
r

some sort of signaling scheme on the computer’s bus to detect and then commu-
nicate with the adapter. Some schemes involve a pulsed electronic signal—almost
fe

like Morse code—sent across the bus, to which the adapter detects and responds.
ea

Such direct manipulation of the bus requires direct hardware access. These types
of programs must then typically be run from DOS and not from within Windows
9x or Windows NT.
Pl
Re

Lesson 6: Data Transmission 149

 Boot PROMs
To support diskless workstations, those without a local hard-disk or floppy-disk
boot PROM: drive, some network adapters support boot PROMs. Because these computers
Chip, used on network don’t have any place to store their operating systems locally, they must retrieve
adapters in diskless them from a server on the network. Computers need a special program to enable
workstations, that stores the them to boot, connect to a server, and download an operating system. Such a pro-

ial
program to boot the gram is typically stored on a special chip on the network adapter.
computer, connect the
network, and download the A boot PROM is a chip that you install (or might come pre-installed) on a net-
operating system. work adapter. The PROM in boot PROM stands for Programmable Read Only
Memory and represents a persistent storage location for data or a program. This
chip stores the program necessary to boot the computer, connect to the network,

er
and download the operating system.
PROM:
Programmable Read Only You must purchase a boot PROM specifically made for your brand and model of
Memory. A persistant storage network adapter. Some newer computers with integrated network adapters come
location for data or a with a boot PROM built in.
program.

at
Drivers

y
A driver is a software program that enables a computer to communicate with a
device. Network drivers are written by the network adapter manufacturers to

op
driver:
A software program that
eM
facilitate communication between their network adapters and the network
enables a computer to redirector on the computer. The job of the redirector is to accept I/O requests for
communicate with a device. files and then send them over the network to the requesting computer. Before a
network adapter can function properly, its corresponding driver must be installed

tC
so that it resides on the computer’s hard disk. Each network operating system
will have its own method for installing drivers. After the driver is installed, typi-
cally there are configuration options, like the ones previously described, that must
be set for the network adapter to function properly. The settings you configure for
No
the driver must match those you set on the network adapter.
c

Many times, manufacturers will change a driver to improve performance. You can
update a driver currently installed on your hard disk with a newer driver. You
should check the manufacturer’s Web site regularly for driver updates. The updat-
en

ing process is typically similar to installing the driver for the first time. If you
Do

remove a device, you should also remove its corresponding driver. This will help
ensure that there will be no conflicts between old drivers and any new drivers
you install at a later time. Again, the removal process is typically similar to the
installation process.
se
r

TASK 6B-1:
fe
ea

Defining network adapter configuration options

Objective: Define the following terms.
Pl
Re

1. Interrupts

2. I/O address

150
3. Base memory address

4. DMA channel

ial
5. Driver

APPLY YOUR KNOWLEDGE 6-2

er
Suggested time:
10 minutes
Identifying adapter and bus types
Examine the various available network adapters and bus types.

at
1. Examine the bus connectors and determine what type of connector it is. Apply Your Knowledge 6–2
is optional.
2. Locate any configuration switches or jumpers.

y
op
3. Examine the bus on the system board and determine the types of slots
available.
eM
4. Determine which of the network adapters could be used with the systems.

Network Adapter Installation tC

The installation of a NIC (Network Interface Card) is similar to that of any inter-
No
face card. You must first know what slot you are going to use for the addition of
c

the network adapter. It is also important to have any necessary tools available for
the install. Part of the Network+ certification is to know what to do in certain
situations. This section will prepare you for situations that involve the installation,
en

configuration or troubleshooting techniques used by technicians when handling

Do

network adapter cards.

The following items can be used as a checklist for things you need to do before
getting inside the PC to install, verify that:
• You have the available bus slot in the machine you wish to add a NIC to.
se
r

• The card is compatible with the protocol, the media, and the network operat-
ing system (NOS).
fe

• You have the available system resources necessary for installation including
ea

open IRQ level, I/O address, and DMA channel.

• The necessary hardware and software has been included. Some items to look
for are drivers, utilities, connectors, terminators or external transceivers.
Pl
Re

Though the physical installation is somewhat easy, you must realize the impor-
tance of doing your research before playing ″the guessing game.″

Lesson 6: Data Transmission 151

 TASK 6B-2:
Installing network cards
Objective: To add a network interface card to your workstation if it
doesn’t already have one, or, if it does have a card installed, to

ial
examine the settings.

1. Does your workstation have a network card installed? If so, continue

with step 5. If not, continue with step 2.

2. Obtain the appropriate type of network card for your system and your

er
network. Turn off your system, open the cover, and prepare to install the
card.

3. Is your network card software or hardware configurable? Install and config-

ure in the appropriate order depending on the card. Using the
appropriate configuration method, configure your card to work with

at
If your network card is your system. This might include setting the IRQ, DMA, I/O ports, and on
hardware configured, you’ll combo cards, the port to use.
need to set the DIP switches

y
or jumpers before installing
it into your system; if your
4. Restart your system. Did Windows detect that you have added hard-
ware? If not, run the Add New Hardware Wizard from Control Panel.

op
network card is software
configured, you’ll need to
eM
Follow the prompts to install the new hardware.
install the card before you
can configure it. 5. Using the Device Manager (Control Panel→System→Device Manager),
check the resource settings for your card. Does the Conflicting Device

tC
List show any conflicts for the card? If so, will the card still work?

6. Return to the Windows desktop.

No
c

Troubleshooting Network Adapters

As a network technician you not only need to implement and maintain the net-
en

work but you also need to troubleshoot. This can be one of the most frustrating
Do

jobs for any network administrator.

Diagnosing the Location of the Problem

se
r

You get a hotline call that a user can’t access the network. Where do you begin
your troubleshooting?
fe

The first thing you need to do is determine where the problem is actually
ea

occurring. Just because a user can’t access the network doesn’t always mean that
the problem is at their workstation. Determine if it is related to a protocol, just a
part of the network, all of the systems on the cable, or the whole network. After
you determine where the problem is, compare the differences between the areas
Pl
Re

experiencing problems and those that aren’t. Try resetting or restarting the
affected devices; this is frequently the solution!

152
If it still isn’t working, try the ″divide and conquer″ method of troubleshooting.
This works particularly well when using a Thin-Ethernet network. Divide the
problem area in half and see if the problem has been cured for half of the area.
Keep dividing the area still experiencing problems until you identify where the
problem is occurring. If you find that the problem is actually the network inter-
face card in the user’s workstation, there are several things you can try.

ial
You can run diagnostic software on the adapter. Sometimes these come with the
card or you might use general-purpose diagnostic software. One test that might be
done is a loopback test. You’ll need a loopback plug to plug into the card if the
card doesn’t have an internal loopback test. Data is sent out and ″loops″ around
back into the card. The ″received″ data is compared to the data that was sent to
verify that it’s the same. If it isn’t, then you’ve got a problem with the network loopback:
Data is sent out and ″loops″

er
card.
around back into the card.
You might also check to see if IRQ, DMA, or I/O port settings have changed. If The ″received″ data is
other equipment was added to the system, the current settings might be conflict- compared to the sent data to
ing with what the NIC is currently set for. verify that it’s the same.
There are internal loopback

at
tests and those that require
an external loopback plug to
APPLY YOUR KNOWLEDGE 6-3 test the card.

y
Suggested time:
Determining why a client can’t access the network 10 minutes

op
eM
In this activity, you need to read through the scenario, then determine what you
would need to do so that the workstation can communicate with the network.

1. You just installed a new NIC in a workstation. This workstation has a

tC
modem on COM1, a mouse on COM2, and a plotter on COM3. There are
no local printers on the workstation. The customer has been unsuccessful in
connecting to the network.
No
c

2. A network user got a second printer and now can no longer access the
en

network.
Do

3. You just added a new system to the network that has a 3C509 combo card
se

installed and you can’t access the network.

r
fe
ea

4. Why should you remove drivers for devices no longer part of your system?
Pl
Re

Lesson 6: Data Transmission 153

 TOPIC 6C
Putting Data on the Cable
In this topic, let’s examine ways that data is put on the data-transmission

ial
medium.

Media Access Control

In network communication systems in which more than one computer has access

er
to the same data transmission media, a standard procedure must be established to
prevent or handle the problem of multiple computers trying to send messages out
on the media at the same time. This procedure is called Media Access Control
(MAC) . If two computers attempt to transmit a signal on the same data-
transmission media simultaneously, the signal from one computer would collide

at
with the signal from the other computer, resulting in both signals being destroyed.
Access methods assure that only one computer at a time can transmit data. They
keep the sending and receiving of data over the network organized. There are

y
four common media-access methods: contention, polling, token passing, and
demand priority.

op
eM
Contention

tC
With the contention media-access method , stations listen to the medium and wait
contention: until no other devices are transmitting before they attempt to transmit a signal. If
A media-access method in another station listening to the medium at the same time attempts to send a mes-
which stations listen until no sage simultaneously, then a collision occurs. When a collision occurs, stations
No
other devices are stop transmitting and wait for a brief period. When the channel is clear again,
c
transmitting before they stations attempt to transmit again.
attempt to transmit a signal.
en
Do
se
r
fe
ea
Pl
Re

Figure 6-14: Contention media-access method.

154
Two types of contention media-access methods are in common use: Carrier
Sense, Multiple Access/Collision Avoidance (CSMA/CA) and Carrier Sense, Mul-
tiple Access/Collision Detection (CSMA/CD).

Contention media-
access method Description AppleTalk uses CSMA/CA
and Ethernet uses CSMA/

ial
CSMA/CA The computer sends a short message that acts as a collision detector. This CD.
signals its intent to transmit data. If no collision occurs with the first signal,
then the computer transmits the actual data.This method increases the
amount of network traffic and slows down performance, because each data
transmission must be preceded with an intent broadcast.
CSMA/CD Each computer on the network checks the data-transmission media for

er
network traffic. If it senses that there is no data being transmitted, it can
send data. Once there is data on the media, no other computer can transmit
until the data has reached its destination and the media is free again.
Collisions are detected and data is re-transmitted if necessary.

at
An advantage to the contention media-access method is that it makes efficient use
of the communication channel in low traffic. However, some disadvantages are:
• It doesn’t guarantee computers access to the data-transmission media. A

y
computer could ″hog″ the channel, blocking the access of other computers.
This access is called probabilistic .

op
•
eM
The flow of data in heavy traffic, because of additional computers on the
network, increases collisions.
• Devices cannot be assigned priorities for faster access.

Polling tC
No
With the polling media-access method , a primary device (also called a controller)
c

controls which secondary device has access to the data-transmission media. The polling:
primary device checks each secondary device in turn to see if it has anything to A media-access method in
transmit. Secondary devices are allowed to transmit for a limited period of time which a primary device
en

before the primary device continues polling. controls which secondary

Do

device has access to the

data-transmissions media.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 155

 ial
er
at
y
op
eM
Figure 6-15: Polling media-access method.
The primary device can give priority (extended access to the data-transmission
media) to certain devices, if necessary. Polling is typically used in hierarchical

tC
networks, in which there is centralized control of network communications.
IBM’s Systems Network Architecture (SNA) is an example of a network that uses
polling. Advantages of the polling media-access method are that computer access
to the data-transmission media is guaranteed, and that there is greater network
No
control due to centralization. A disadvantage of the polling media-access method
c

is that it doesn’t use the communication channel efficiently in low traffic; instead,
it ties up excessive bandwidth while listening for and acknowledging messages.
en
Do

Token Passing
Token passing is similar to polling, except that a master station doesn’t control
token passing: access to the data transmission media. Instead, network computers pass a token ,
A media-access method in a special sequence of bits that grants the station holding it permission to transmit
se

which the station with the data. If a station doesn’t need to transmit, it passes the token and the data on to
r

token is the only one who the next computer on the network, which passes it on to the next computer, and
can transmit. so forth, until eventually the token and the data returns to the transmitting
fe

computer.
ea
Pl
Re

156
 ial
er
at
y
op
eM
Figure 6-16: Token passing media-access method.
The transmitting computer checks for transmission errors by comparing the data
it received with the data it transmitted. It then releases the token to the next

tC
computer. In this manner, every computer has an opportunity to transmit.
Token passing is a deterministic media-access method. That is, each computer is
guaranteed access to the data transmission media. This is ideal for networks in IBM’s Token Ring and FDDI
which timing is critical, as in automated manufacturing. Advantages of the token- use the token passing
No
passing media-access method include: it grants guaranteed computer access to the media-access method.
c

data transmission media; it eliminates collisions; and it is efficient under high

traffic conditions. Disadvantages of the token passing media-access method are: it
provides inefficient use of the data transmission media during low traffic; all
en

devices use complex software requiring intelligence; and device software requires
Do

adjustment when you add or eliminate other devices.

Demand Priority
se

Demand priority is a relatively new access method designed for a specific net-
r

work type called 100VG-AnyLAN. In demand priority, the hub determines when demand priority:
each computer can transmit on the data-transmission media. If a computer wants A new media-access method
fe

to transmit, it signals the hub. Such a signal is called a demand . The hub returns for 100VG-AnyLAN networks.
ea

a signal giving permission to transmit one small portion of the data, called a A hub determines when each
packet . When multiple computers want to transmit, the hub cycles through each computer can transmit.
of the requesting computers in port order. Each computer is allowed to transmit
Pl

one packet per cycle. Computers that haven’t signaled the hub that they want to
Re

transmit are skipped in the cycle and don’t take time in the transmission cycle.

Lesson 6: Data Transmission 157

 TASK 6C-1:
Acting out media-access methods
Objective: Identify which media-access method is being described and
identify any drawbacks of each of the media-access methods

ial
described in each of the scripts.

1.
COMPUTER 01: I’m ready to send. I see the data transmission media
is free, so…

er
COMPUTER 01: Hi, COMPUTER 02! How are you today? I’m done
sending…
COMPUTERS 02 and 03: I’m ready to send. Whoops! Someone else is
sending, so I’ll wait.
[Pause]

at
COMPUTERS 02 and 03: I’m ready to send. Whoops! Someone else is
sending, so I’ll wait.

y
[Pause]
COMPUTER 02: I’m ready to send. I see the data transmission media

op
eM
is free, so...
COMPUTER 02: Hi, COMPUTER 01! I’m doing fine, thanks. I’m done
sending.

tC
No
c

2.
en
Do

MEDIATOR: COMPUTER 01, do you have anything to say?

COMPUTER 01: No.
MEDIATOR: COMPUTER 02, do you have anything to say?
COMPUTER 02: Yes, as a matter of fact. Fourscore and...
se
r

MEDIATOR: Your time is up. You’ll have to finish later. COMPUTER

01, do you have anything to say?
fe

COMPUTER 01: No.

ea

MEDIATOR: COMPUTER 02, do you have anything to say?

COMPUTER 02: Yes. (Where was I?)...seven years ago, our...
MEDIATOR: Time’s up. COMPUTER 01, do you have anything to
Pl
Re

say?

158
 COMPUTER 01: No.

ial
3.
COMPUTER 01: I’ve got the token. (Ahem!) Fourscore and ...
Whoops! My time’s up. Here’s the token, COMPUTER 02.
COMPUTER 02: Thanks, but I don’t need it. Here’s the token, COM-
PUTER 03.

er
COMPUTER 03: Thanks, but I don’t need it either. Here’s the token,
COMPUTER 01.
COMPUTER 01: Thanks. I’ve got it. (Where was I?)...seven years ago,
our... Whoops! Where does the time go? Here’s the token COMPUTER
02.

at
COMPUTER 02: Thanks, but I still don’t need it. Here’s the token,
COMPUTER 03.
COMPUTER 03: Thanks, but I don’t need it either. Here’s the token,

y
COMPUTER 01.

op
eM
TOPIC 6D tC
No
c

Network Topologies
A topology is a description of the layout of a specific region or area. A network
en

topology is a description of the layout of the region or area covered by a network

Do

. The complete physical structure of the cable (or data-transmission media) is

called the physical topology . The way data flows through the network (or data-
transmission media) is called the logical topology . While the physical topology
might be classified as part of the OSI Physical layer, it is constrained and typi-
cally defined by the logical topology of a network. Since the logical topology is
se

part of the Data Link layer, the network topology is typically considered part of
r

the Data Link layer.

fe

There are two types of connections that describe how multiple devices connect to
ea

a single cable or segment of data-transmission media. They are point-to-point and

multi-point. Point-to-point connections provide a direct link between two devices;
for example, a computer connected directly to a printer. Multi-point connections
provide a link between three or more devices on a network. All computer net-
Pl
Re

works, regardless of their topologies, rely on point-to-point and multi-point

connections.

Lesson 6: Data Transmission 159

 Common Topologies
Let’s examine some of the common topologies you might encounter on a
network.

ial
Bus Topologies
A bus topology uses one long cable (backbone) to which network devices are
either directly attached or attached by using short drop cables. Because all com-
puters share this bus, a computer checks for any information that might be
coming down the backbone before sending its message. A message passes the
other computers on the way to its destination. Each computer then checks the

er
address of each message to see if it matches its own. Bus topology is a passive
topology , which means that computers connected to a bus topology only listen
for a signal being transmitted over the data-transmission media. They don’t move
data from one computer to the next. Thus, if one computer goes down, it doesn’t
affect the rest of the network.

at
y
op
eM
tC
No
c
en
Do
se
r

Figure 6-17: Bus network topologies.

In a bus network topology , the backbone must be terminated at both ends to pre-
fe

vent the signal from echoing back when it has reached the end of the bus. A
ea

terminator is a resistor that essentially absorbs signals when they reach the end of
the cable. Thus, a terminator is said to have the effect of making the cable look
infinite to a signal. Typically, with shielded cable installations, such as with
coaxial cable, one of the terminators is connected to ground so as to dissipate
Pl
Re

static electricity that might build up in the shielding of the cable.

The advantages and disadvantages of a bus topology are listed below:

160
Advantages Disadvantages
Uses much less cable than other topologies use. A break in the data transmission medium adversely
affects the entire network.
Individual computers can fail without taking down Difficult to troubleshoot and reconfigure.
the entire network.
Easy to install.

ial
Ring Topologies

er
Ring topologies consist of several computers joined together to form a circle.
Messages move from one computer to the next in one direction only. When a
computer receives a message addressed to it, the message is copied and sent back
with a modification to indicate that it was received. Ring topology is an active
topology , which means that computers connected to a ring topology are respon-
sible for moving data from themselves to the next computer. Thus, if one

at
computer goes down, it affects the rest of the network.

y
op
eM
tC
No
c

Figure 6-18: A ring topology.

en
Do

Some networking schemes use a dual ring topology . In these configurations, data
moves in opposite directions around the two rings. In most cases, one ring is the
primary route over which data flows; the second ring serves as a backup in case
of a break in the first ring. In the event of a primary ring break, the primary and
secondary rings are automatically connected to form one larger ring. A single ring
se
r

break won’t bring down a dual-ring network. Such dual ring configurations are
called dual counter-rotating rings , because when un-joined, data flows in the
opposite direction on the secondary ring as that on the primary ring.
fe
ea

Advantages and disadvantages of ring topologies are as follows:

Advantages Disadvantages
Pl

Almost no loss in signal quality over the network Failure of a single connection can take down the
Re

because signals are transmitted by each computer. entire network (unless the more expensive dual-
ring topology is used).
Easy to install. Difficult to reconfigure.
Easy to troubleshoot.

Lesson 6: Data Transmission 161

 Advantages Disadvantages
Topology is unidirectional and continuous (that is,
there is no end of the network to reflect signals).
Reflected signals aren’t an issue.

ial
Star Topologies
The star topology can be used in either Ethernet or Token Ring networks. In
Ethernet, the star topology is referred to as a star-bus. It uses a central device,
called a hub , with drop cables extending in all directions. Each networked device

er
is connected point-to-point to the hub. Hubs can be active or passive . The major-
ity of hubs are active, meaning they regenerate and retransmit the signal. Some
hubs, such as wiring panels, are passive, meaning they act only as connection
active hub: points and don’t regenerate the signal. All messages in a star topology must go
Hubs that regenerate and through the central hub before reaching their destination. Star topology is a pas-

at
retransmit the signal. sive topology. Computers connected to a star topology aren’t responsible for
moving data from one computer to the next. Thus, if one computer goes down, it
doesn’t affect the rest of the network. In Token Ring, the star topology is referred

y
to as a star-ring. It connects workstations via patch cables. MSAUs are connected
passive hub:
together to create the ring.

op
Hubs that act only as a
connection point and do not
eM
regenerate the signal.

tC
No
c
en
Do
se
r
fe

Figure 6-19: A star topology.

ea

Just because a network connects to a central networking device does not mean
it’s a star topology network. Take for instance 10BaseT Ethernet. Like all
Ethernet networks, 10BaseT Ethernet uses a bus topology. However, with
Pl

10BaseT Ethernet, twisted-pair cables run from computers to hubs giving the
Re

appearance of a star topology. Inside the hub, however, each of the cables is con-
nected forming the bus required for Ethernet. The star appearance is just an
illusion.

162
Advantages and disadvantages of star topologies are as follows:

Advantages Disadvantages
de jure standard:
Inexpensive. Low data rate. A standard that exists
Easy to troubleshoot. No de jure (mandated) standards. through codes, laws,
Easy to reconfigure. Moderately difficult to install. decrees, or other forms of

ial
legislation. (Compare with de
Require more cable than other topologies. facto standard.)

de facto standard:
Mesh Topology A standard that exists

er
In a mesh topology network , every computer is directly connected to every other through popular practice.
computer. This topology is rarely if ever used in the real world, though some
incorrectly believe the Internet is a mesh topology network. In a mesh topology
network with an n number of computers, you must have (n * (n-1))/2 two-way
connections. Thus, to connect four computers (n=4) you must have 6 connections

at
(4 * (3))/2 = 6; with eight computers, you must have 28 connections; and with
100 computers, you must have 495 connections. You can probably see why this
method is not frequently used.

y
op
eM
tC
No
c
en
Do
se
r

Figure 6-20: A mesh topology.

fe

Advantages Disadvantages
ea

Every computer is connected to every other A potentially enormous number of connections for
computer, virtually assuring constant networks of even modest size makes mesh networks an
networking communications. unreasonable venture. The many connections necessary
Pl

would lead to exorbitant costs and exceedingly difficult

Re

management burdens.

Lesson 6: Data Transmission 163

 Hybrid Topologies
A hybrid topology is a combination of two or more different topologies. WANs
sometimes have hybrid topologies because they connect a variety of LAN
topologies. Two common hybrid topologies are the star bus and the star ring. In a
star bus topology, there are several star topology networks connected together via
their hubs in a bus topology. The star ring topology is similar to the star bus

ial
topology. However, instead of having the hubs connected together in a linear bus
topology, they are connected to a main hub in a star topology. The big advantage
of hybrid topologies is that they connect disparate topologies. The disadvantage
of hybrid topologies is that they are potentially complex to establish and
maintain.

er
at
y
op
eM
tC
No
c

Figure 6-21: A hybrid topology.

en
Do
se
r
fe
ea
Pl
Re

164
TASK 6D-1:
Identifying Topologies
Objective: Identify each of the topologies illustrated.

ial
1.

er
at
y
op
eM
2.

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 165

 3.

ial
er
at
y
op
eM
4.

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

5. What type of network topology is used in your classroom? What is used

at work?

166
Choosing a Physical Topology
The following characteristics are important to consider when choosing a physical
topology:
• Relative ease of installation, reconfiguration, and troubleshooting.
• The maximum number of units affected by a data-transmission media failure.

ial
The following table compares the characteristics of the bus, ring, and star topolo-
gies:

Bus topology Ring topology Star topology

Overview Because it uses Moderately easy to Installation is moderately

er
established standards, it install, but you are difficult. The design of the
is relatively easy to limited by the network is simple, but you
install. maximum ring length must install a separate
and the number of data-transmission media
devices. for each computer or
device.

at
Installation Installation typically Installation typically Installation requires more
requires less data- requires more data- cabling than most other
transmission media transmission media topologies.

y
(cable) than other than the bus topology.

op
topologies.
Reconfiguration Reconfiguration is
eM
Reconfiguration Reconfiguration is
difficult, especially when becomes more difficult relatively easy. Moves,
distance or number of as the scale of additions, or changes don’t

tC
taps are at the maximum relocation increases. involve more than the
allowable levels. connection between the
computer or device and the
central computer.
Troubleshooting Troubleshooting is Because each device Troubleshooting is
No
difficult, because all incorporates a repeater relatively easy because all
c

computers or devices on (a device that data in a star topology

the network might have regenerates a signal), goes through a central
to be tested to locate the you can easily find point, where it can be
en

problem. cable faults. The distributed to the

Do

downstream repeater appropriate address.

from a failure identifies
the error when it
doesn’t receive a signal
from an upstream
se

repeater within a
r

specified amount of
time.
fe

Media failure All computers or devices Data-transmission Data-transmission media

ea

affects are affected by a data- media failure on a faults are automatically

transmission media single loop system isolated to the failed
failure; failure of a single affects all devices on segment.
computer doesn’t affect the network. However,
Pl
Re

the rest of the network. if you have several

loops, only the
computers or devices
on the affected loop
will go down.

Lesson 6: Data Transmission 167

 TASK 6D-2:
Identifying advantages and disadvantages of network
topologies
Objective: Identify the topologies described in each step below.

ial
1. Installation typically requires less data-transmission media (cable) than
all other topologies.

er
2. All data goes through a central point, where it is distributed to the
appropriate address.

3. Uses one long cable to which network devices are either directly

at
attached or attached by using short drop cables.

y
4. Several computers or devices are joined together to form a circle.

op
eM
5. Messages move from one computer or device to the next only in one
direction.

6. tC
Messages pass through all other computers on the way to their
No
destinations.
c

7. Uses a central hub with drop cables extending in all directions.

en
Do

Suggested time:
APPLY YOUR KNOWLEDGE 6-4
10 minutes
se

Examining network topologies

r

In this activity, you’ll view the Network Topologies presentation on the Network+
fe

Certification CD-ROM.
ea

1. Access the Digital Video component of the Network+ Certification

CD-ROM.
Pl
Re

2. Click on the Network Cables And Devices link, then the Network Topologies
link.

3. Click Overview to begin the presentation.

4. When finished, exit from the Network+ Certification CD-ROM.

168
TOPIC 6E
Bridges
Repeaters amplify the signals they receive and just send them on; they don’t

ial
check the packet to see where it’s headed. Bridges, on the other hand, do read the
packet to find the MAC address. It can then filter the traffic based on the destina-
tion MAC address.
A bridge of any kind—transparent, source-routing, or source-routing transparent
(SRT)—increases the amount of data that passes through a LAN by determining
which frames have a destination on the local network and which should continue

er
on to other LANs. It keeps those that should stay and passes the rest on, much
like a filter system. The frame is then repeated along the next segment toward its
destination. This process extends the transmission media’s length and device
limitations.

at
Transparent, source-routing, and source-routing transparent bridges are each dis-
cussed in the following sections of this topic.

y
80/20 Rule

op
eM
The 80/20 rule is simply a guideline designed to help create an optimally efficient
bridged network. It suggests that 80 percent of your network traffic should be
local, leaving only 20 percent of it to be internetwork. If traffic is local, two seg-

tC
ments connected by a bridge can be used at the same time once the bridge
discovers the receiver’s location. Because of this, it is more efficient and produc-
tive to have most of your network traffic be local.
You could use a repeater, but when doing so, frames can be transmitted on only
No
one segment at a time because they are simultaneously repeated on the other seg-
c

ment, as shown in Figure 6-22.

en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 169

 ial
er
at
y
op
eM
Figure 6-22: The 80/20 rule in regard to bridges and repeaters.

Transparent Bridge tC
Requiring little or no setup, a transparent bridge is invisible to a transmitting
No
device. The sender does not know that its data might have to travel through one
c

or more bridges to a remote segment. For that reason, this type of bridge was
aptly named ″transparent.″ In regard to forwarding data, a transparent bridge per-
forms typical bridge functions—it checks the target device’s MAC address and
en

forwards the data to the appropriate segment.

Do
se
r
fe
ea
Pl
Re

170
 ial
er
at
y
op
eM
Figure 6-23: A frame passing through a transparent bridge.

every segment on a network except the segment from which the frame came. By
tC
Upon establishing a new transparent bridge, it sends every frame it receives to

using a filtering database , the bridge ultimately learns to filter frames. A filtering
database contains the network locations of end nodes.
No
c

Filtering Database
en

The database contains a few permanent entries, such as broadcast addresses,

Do

which are entered by the vendor. Additional entries are made either manually by
the administrator or dynamically as the bridge watches (learns) network traffic.
Eventually, the database contains an address entry for each network device, as
well as each broadcast and multicast address that will be passed between
segments. All of this information is required because a filtering database uses a
se
r

flat addressing scheme.

During its learning process, a transparent bridge looks at a frame’s sender address
fe

and discovers which segment each network devices lives on, as shown in Figure Transparent bridges are also
ea

6-24. Once it learns where each device is, it forwards only those frames that are referred to as learning
traveling to devices on remote segments. bridges because of the way
they ″learn″ network traffic.
Pl
Re

Lesson 6: Data Transmission 171

 ial
er
at
Figure 6-24: Creating a filtering database.

y
The connection point between a bridge and a media segment is called a bridge

op
Spanning tree protocol is
eM
port . A bridge port exists in one of five modes, called port states . The bridge
discussed later in this topic. protocol (spanning tree) or management entity (SNMP agent) tells the bridge port
which state it should be in. When a bridge port is spanning tree, it moves through
the port states, excluding the Disabled state, which does not participate.

A bridge port is similar to a

network adapter in that it •
tC
The five port states, in order, are Disabled, Blocking, Listening, Learning, and
Forwarding.
Disabled. The port does not learn, and is not included in spanning tree
No
attaches to a media segment. operation forwarding. This state is used in bridge management as a way to
c

manually take the bridge port offline.

• Blocking. The bridge is in standby mode. A bridge is in this state when first
started or if it is a backup in a redundant bridge pair. It processes only
en

bridge protocol frames sent to the bridge multicast address.

Do

• Listening. The port is getting ready to learn and forward (the next two
states). It can be in this state only until the confusion of a topology change
dies down and the network is again ″settled.″ During this time, frames are
not forwarded and learning does not happen so that incorrect information
se

will not be added to the filtering database and bridging loops.

r

• Learning. The port can add information that it gathers to the filtering
database. However, it cannot forward frames yet. Like the Listening state,
fe

there is a time limit at this phase that gives the network time to settle down
ea

before frames are forwarded. This prevents unnecessary frame forwarding

and temporary bridging loops.
• Forwarding. The typical, normal port state. It can both learn and forward
Pl
Re

frames. This is the only state in which forwarding occurs.

172
 ial
er
at
y
op
eM
Figure 6-25: The five transparent bridging port states.

Relay Entity
tC
A bridge will not automatically forward all frames. For example, a bridge cannot
forward a frame addressed to the bridge itself. In order for a frame to be for- Transparent bridging
No
warded, it must have a valid Cyclic Redundancy Check (CRC), be traveling to a happens at the MAC
c

remote segment that is included in the filtering database, and include data from sublayer, so it considers
upper layers (the LLC sublayer and up). anything upwards of itself to
be upper-layer.
A service known as the relay entity performs forwarding and filtering for a
en

bridge, provided that the bridge port is in the Forwarding state. The relay entity
Do

uses the filtering database to figure out whether to forward or drop a frame.
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 173

 ial
er
at
y
op
Figure 6-26: The relay entity.
eM
While a frame’s CRC is being verified, a transparent bridge reads and holds onto
the message, a technique called store-and-forward. To verify the CRC, the bridge
checks the frame’s length against the value specified within the frame. If the

tC
CRC checks out, the bridge forwards the message.
Store-and-forward mode has advantages and disadvantages. It stops corrupted
It’s important not to confuse frames from being produced on the network, but creates delays or latency, which
increase proportionally with the size of a frame.
No
this type of store-and-
forward process with that of
c

a store-and-forward network;
a store-and-forward network
is a message switching Bridging Loops
en

process that is used for

A bridging loop occurs when redundant bridges are used to give fault tolerance to
Do

delivering email.
a bridged LAN. A bridging loop is a continuous circular exchange of data packets
between two redundant bridges in which the packets are never transmitted to their
destinations. Having redundant paths in a transparent bridge environment can
cause these potentially fatal loops. A broadcast storm (where new packets are
continuously produced by redundant bridges) that occurs when the spanning tree
se
r

protocol is not in use is also referred to as a bridging loop. The following figures
and scenarios explain how bridging loops occur.
fe
ea
Pl
Re

174
 ial
er
at
y
op
eM
Figure 6-27: How a bridging loop is created.
Redundant bridges 1 and 2 connect segments 1 and 2. As Node 1 sends data to

tC
Node 2, Bridge 1 and Bridge 2 each copy the frame from segment 1 and add an Remember that bridges
entry for Node 1 to their filtering databases. The bridges are not physically con- always use the most current
nected to each other, so one bridge forwards the frame before the other. In Figure information.
6-28, Bridge 1 forwards the frame before Bridge 2 does.
No
c
en
Do
se
r
fe
ea
Pl
Re

Figure 6-28: Bridging loops, part 2.

Lesson 6: Data Transmission 175

 Bridge 2, not knowing that Bridge 1 sent the frame, copies it from segment 2 and
checks its filtering database. When it sees that it already has an entry for Node 1
on segment 1, it thinks the information is outdated and changes it to reflect Node
1 on segment 2. In the meantime, Bridge 1 copies the original frame forwarded
by Bridge 2 to segment 2 and thinks its filtering database is outdated. It changes
the database to reflect that Node 1 lives on segment 2 and forwards the frame to
segment 1.

ial
The same situation occurs on segment 1 and the frame is cycled back to segment
2, and so on between the two segments. The frame is never forwarded to its true
destination, Node 2 on segment 2. This happens to all frames between both seg-
ments; eventually, LAN traffic becomes so heavy that the LAN cannot function.

er
TASK 6E-1:
Discussing transparent bridges

at
1. Explain the 80/20 rule.

y
op
eM
2. How do a transparent bridge’s functions differ from those of a typical
network bridge?

tC
No
3. What service filters and forwards data for a transparent bridge?
c

4. You have two redundant bridges on your LAN for fault tolerance. Data
en

is sent from one node to another but never arrives at its destination.
Do

What might have happened? Why?

se
r
fe
ea
Pl
Re

176
CHECK YOUR SKILLS 6-2 Suggested time:
2 minutes
Identifying bridge port states
Identify the following phrases with the appropriate bridge port state.

ial
1. Standby mode

2. Normal state

er
3. Offline

at
4. Unsettled network

y
5. Gathering, but no forwarding

op
eM
Spanning Tree Protocol

tC
The spanning tree protocol arranges segment routes and redundant bridge routes
so that bridging loops do not occur. It is a bridge hierarchy protocol defined by
IEEE 802.1d that includes a root bridge at the top of the tree, and designated
spanning tree protocol:
Arrangement of segment
No
bridges below it. The spanning tree protocol declares one of a pair of redundant routes and redundant bridge
c

bridges the designated bridge, and the other, the backup. This enables the bridge routes to prevent brigding
to identify routes that contain no bridging loops. It also enables redundant bridges loops.
to communicate with each other so that route failures are detected and bridge
en

routing can be dynamically reconfigured around the faulty bridge or segment.

Do

The root bridge is responsible for letting all bridges in the spanning tree know if
something has changed on the network. It periodically issues configuration mes-
sages to all designated bridges from which they can determine whether they need
to reconfigure their routes. The designated bridges forward the messages.
se
r

Spanning Tree Configuration

fe
ea

The method by which the spanning tree protocol enables bridges to communicate
is a packet called the Bridge Protocol Data Unit (BPDU) . The bridges use
BPDU to dynamically choose a root bridge, designated bridges, and backup
bridges for routes between segments.
Pl
Re

Lesson 6: Data Transmission 177

 A spanning tree network contains only one root bridge, which is always the net-
work bridge with the lowest bridge ID (bridge identification number) . Bridge
IDs can be randomly assigned; however, it is best if you manually assign them. A
bridge ID is an 8-byte number of which you assign the first two bytes and specify
the last six bytes to be the assigned port adapter hardware address, as displayed
in Figure 6-29.

ial
When buying a bridge, it’s a
good idea to look into its
management implementation.

er
Figure 6-29: A bridge ID.
It is a good idea to choose a bridge that is physically closest to the network’s

at
center as the root bridge. (Assign that bridge the lowest two-byte number in its
bridge ID.) This decreases the time it takes to reconfigure the network in the
event of segment or bridge failures.

y
op
eM
Root Bridge Selection
Selecting a root bridge starts with all bridges broadcasting a frame to all of its
ports. The frame contains the following fields: Dest Addr, Srce Addr, DSAP,

tC
SSAP, BPDU, CRC, MAC Header, LLC Header, BPDU, and CRC, as displayed
in the Figure 6-30.
No
c
en
Do
se
r
fe
ea

Figure 6-30: Fields included in the root selection frame.

Pl
Re

The BPDU section of the frame consists of constants, flags, a root ID, port cost, a
bridge ID, and other miscellaneous items.

178
 ial
er
at
Figure 6-31: A breakdown of the frame’s BPDU section.
The following table is a breakdown of the BPDU section’s components.

y
Section Component(s) Size Description

op
Constants Protocol identifier 2 bytes
eM
Version 1 byte
Message type 1 byte

tC
Flags TC 1 byte Topology Change notification that is used to
adjust timers during a reconfiguration.
TCA 1 byte Topology Change Acknowledgment.
Root ID 8 bytes The root bridge’s identification number.
No
Port Cost 4 bytes The cost of the route to the root.
c

Bridge ID 8 bytes The ID of the bridge that is sending the

message.
Misc Port ID 2 bytes The determining factor if two bridges have
en

equal bridge IDs.

Do

Message age 2 bytes The amount of time between configuration

messages.
Max age 2 bytes The time after which a message should be
purged.
Hello time 2 bytes The amount of time since the last
se
r

configuration message from the root.

Forward delay 2 bytes The amount of time between states during a
fe

topology change.
ea

Each bridge then puts its bridge ID into the Bridge ID and Root ID fields of the
BPDU field, indicating itself as the root bridge. Each frame’s LLC Control field
Pl

then instructs each bridge to copy all incoming BPDUs. When the bridge does so,
Re

it checks the Root ID field. If the field contains a bridge ID lower than its own, it
stops broadcasting its BPDUs and forwards the BPDU with the lower-bridge ID.

Lesson 6: Data Transmission 179

 ial
er
at
y
op
eM
Figure 6-32: The root bridge selection process.
In Figure 6-32, Bridge 2 has the lowest bridge ID and becomes the root bridge.
Bridge 1, by receiving BPDUs from Bridge 2, realizes that it is parallel to Bridge
1 and becomes Bridge 2’s backup.

tC
Once the root bridge is selected, it sends BPDUs from all of its ports every
couple of seconds by using spanning tree protocol. (IEEE recommends a two-
second default.)
No
c

Designated Bridge Selection

en

Each bridge copies the BPDUs received through its root port (a bridge port facing
Do

the root bridge). The BPDU sent by the root bridge has a field called Port Cost,
which the root bridge automatically sets to 0. (At this point, you have already
assigned all other bridge ports a port cost value.)
The bridges consecutively receive the root bridge’s BPDU and add their config-
ured port costs to the port cost value sent by the root. They copy each other’s
se
r

BPDUs from the segment and check the port costs. The bridge with the lowest
port cost is the designated bridge; the bridge with the higher port cost is the
backup.
fe
ea
Pl
Re

180
Bridge or Segment Failure
If a bridge does not receive a BPDU from the preceding bridge within a certain
amount of time, it assumes that the bridge is malfunctioning and sends a Topol-
ogy Change Notification (TCN) from its root port. This is the only time a bridge
will send BPDUs from its root port after a root bridge has been selected. The
bridge continuously sends the TCN on the root port until the designated bridge

ial
acknowledges it upstream. Eventually, the root bridge receives the TCN and
sends a Topology Change Acknowledgment (TCA) in the BPDUs it sends every
two seconds.
Upon receiving the BPDU from the root bridge, each bridge deletes all non-
permanent entries from its filtering database and resumes operation in the

er
Blocking state. The root bridge sends a BPDU specifying how long the bridges
should remain in this state. The amount of time is set by the administrator and
must be sufficient for all bridges to receive the BPDU containing the topology
change.
After the specified amount of time has passed, the bridges repeat the designated

at
bridge selection process. The new route around the failed bridge or segment is
automatically configured and the bridges, once again, go through each state
(Blocking to Forwarding) after which they restart forwarding frames.

y
op
Spanning Tree Protocol Guidelines
eM
The following list is a set of guidelines to follow when configuring the spanning
tree protocol:
•
•
•
Have redundant routes to every segment.
Do not make frames travel through more than two bridges.
tC
Do not make frames travel through more than three bridges after a topology
No
change (where the backup bridge becomes the designated bridge).
c

An efficiently configured spanning tree protocol should provide you with fast
topology changes, minimal reconfiguration time for those changes, and minimal
en

latency. For example, BPDUs are forwarded only one level from the root bridge,
Do

so topology changes can happen almost instantly. TCNs from backup bridges
have to travel over only one segment to reach the root bridge, so reconfiguring
after a topology change happens quickly. Also, having no more than three bridges
in a route reduces latency.
se
r

TASK 6E-2:
fe
ea

Discussing the spanning tree protocol

1. How does the spanning tree protocol help prevent bridging loops?
Pl
Re

Lesson 6: Data Transmission 181

 2. During root bridge selection, which network bridge becomes the root
bridge?

3. During designated bridge selection, which bridge becomes the designated

bridge?

ial
4. When configuring the spanning tree protocol, do you want to have
redundant routes to each segment?

er
5. What is the minimum number of bridges frames should travel through?

6. What are some benefits of the spanning tree protocol?

at
y
op
eM
Source-routing Bridge
Source-routing bridges differ from transparent bridges in that they do not use fil-
tering databases. Instead, they use dynamic routing tables, which contain paths to

tC
each device a source-routing bridge communicates with. Source-routing bridges
are typically used in Token Ring networks.
The frame that a device sends out contains routing information in its MAC
header. The source-routing bridge uses this information to forward or drop the
No
frame. Unlike transparent bridges, source-routing bridges copy only frames with
c

source-routing information in their headers. (A transparent bridge copies all

frames from a segment.) Even though frames on a source-routing network contain
routing information, they use a bridging process and not a routing process,
en

because they use physical addresses.

Do
se
r
fe
ea
Pl
Re

182
 ial
er
at
Figure 6-33: Frames using a source-routing bridge.

y
In Figure 6-33, Node 1 transmits data to Node 2. In its MAC header, Node 1

op
eM
includes the route by which the data should travel to Node 2 on segment 3.
Bridges 1 and 2 use that information to forward the data.

Route Discovery
The means by which a source-routing device creates a routing table is called tC
route discovery. Route discovery is a process that establishes a route table based
No
on a hello frame between a sending and receiving device. The sending device
c

transmits the hello frame to the receiving device. As the frame travels to the
receiver, each bridge it passes through adds route information to the frame’s
existing route information, dynamically creating a route table.
en
Do

The table is updated when the receiving device responds to the sending device
with the completed information. The sending device adds it to the routing table.
Route tables are dynamic and are usually stored in RAM. When a device’s power
is turned off, route table entries are cleared from the table. If the bridge is not
turned off, entries stay in the table until the station entry is reset or overwritten
when the table limit is maximized.
se
r

In a Token Ring network, devices minimize traffic by attempting to find a device

locally. A frame is first sent out without source-routing information. If it returns
fe

with the Address Recognized Indicator bit specified, the sending device knows
ea

that the receiving device is on the local ring.

Many source-routing networks, such as IBM, use LLC eXchange Identification
(XID) packets as hello packets, but you can use any hello-oriented protocol that
Pl
Re

creates establishes communication between two systems.

Lesson 6: Data Transmission 183

 TASK 6E-3:
Discussing source-routing bridges
1. What is the main difference between source-routing and transparent
bridges?

ial
2. Can route tables be stored in RAM?

er
3. What is the purpose of a hello packet?

at
Suggested time:
APPLY YOUR KNOWLEDGE 6-5
30 minutes

y
Reviewing the Physical layer topic

op
eM
In this activity, you’ll work through some sample test questions that deal with the
information presented in this (and the previous two) lessons.

1. Load the Skill Assessment component of the Network+ Certification

tC
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ text.

3. Select all items for the Physical Layer topic, uncheck the Test Time check
No
box, and then click OK.
c

4. Work through the sample questions and score the test.

en

5. If you want to review the questions and correct answers, click Review Items.
Do

Use the Explanation and Next buttons to work through the questions and
answers; then use the Scores button to return to your test score window.

6. Exit the test.

7. Start the Network+ test again, this time choosing the Data Link Layer topic
se
r

and uncheck the Test Time check box; then click OK.
fe

8. Review the questions and answers as described in step 5.

ea

9. Exit the test.

Pl
Re

184
 Summary
In this lesson, you’ve examined network connection devices, the role of net-
work adapters in network communication, various access methods for
getting data on the transmission medium, and identified characteristics of

ial
various network topologies. You also examined the use of bridges in
networking.

LESSON 6 REVIEW

er
6A Explain the purpose of network connection devices and describe issues
associated with them.

at
y
6B What is the role of a network adapter?

op
eM
6C What function does an access method provide for data transmissions?

tC
No
c

6D Draw an illustration of each of the following topologies: Bus

en
Do
se
r
fe
ea
Pl
Re

Lesson 6: Data Transmission 185

 Star

ial
er
at
Ring

y
op
eM
tC
No
c
en

6E How does a transparent bridge forward data packets?

Do

What is a bridging loop and how does it occur?

se
r
fe
ea

How do source-routing bridges differ from transparent bridges?

Pl
Re

186
The Network Layer LESSON
7

ial
Overview Data Files:
Routing occurs at the Network layer of the OSI model. Routers and brouters Network+ Certification
are used to link subnetworks (and separate networks) together. Only some of CD-ROM

er
the network protocols in use on networks are routable protocols; you’ll
examine which are and which aren’t. You’ll also take a look at the use of Lesson Time:
default gateways and subnetworks, and the use of unique network IDs. 2 hours
Finally, you’ll examine the differences between static and dynamic routing.

at
Objectives

y
To understand the function of the Network layer of the OSI Model, you will:

op
7A
eM
Describe addressing associated with the Network layer.
You’ll see how addresses are used to route information to the appropriate
network.

7B Explain the purpose of network connection devices and describe

issues associated with them.
tC
In this topic, you’ll examine several internetworking devices and see
No
when they are appropriate. For example, if you’re using a non-routable
c

protocol, you won’t want to use a routing device.

7C Describe the various components associated with routing, including

en

route tables, protocols, and devices.

Do

This topic includes information on routers, routable and non-routable pro-

tocols, default gateways, subnetworks, unique network IDs, and static and
dynamic routing.
se
r
fe
ea
Pl
Re

Lesson 7: The Network Layer 187

 TOPIC 7A
Addresses
Networks, especially large internetworks, might provide more than one path for

ial
moving information from one point in the network to another. The Network lay-
er’s basic purpose is to decide which physical path the information should take to
move from its source to its destination. Determining the path to take is called
routing. The path might span the distance between two computers in the same
office or it might span the globe.
The Network layer of the OSI model might appear similar to the Data Link layer,

er
because these two layers are concerned with moving data to the proper location.
The difference is that the Data Link layer’s concerns and addressing issues were
While the Data Link layer confined to a single segment, or link, in the network, whereas the Network layer’s
deals with issues on a single concerns and addressing issues span networks.
segment of the network, the

at
Network layer deals with The network layer:
issues spanning networks.
• Moves information to the correct address.
• Assembles and disassembles packets.

y
• Addresses and routes data packets.

op
•
eM
Determines the best path for moving data through the network.

tC
Logical Address
The manufacturer usually assigns MAC addresses to devices at the time the
devices are built. The administrator, on the other hand, usually assigns Network
layer addresses. This network address is a logical address, sometimes referred to
No
as the address of the cable. This address is of primary concern to Network layer
c

addressing.
The Network layer address distinguishes one network from another, enabling
en

packets to be routed to their ultimate destination and filtered so that they don’t
Do

get forwarded to inappropriate networks. To be able to route packets, all logical

addresses must be unique. To ensure uniqueness across multiple networks, a com-
bination of logical (network) and device (physical) addresses is used.
se
r
fe
ea
Pl
Re

188
Service Address
Some devices can perform several functions simultaneously. These collective
functions are known as entities. An entity must have a unique address so that it
can transmit data. This identifying address is referred to as a service address and
is a combination of a device’s logical, physical, and service addresses. The ser-
vice address is appended to the logical and physical addresses, enabling an entity

ial
to communicate. A service address is sometimes referred to as a port or socket service address:
address. Address (number) of a port
or socket.
In the event that two network processes from different logical networks might
have the same service address, the rule for unique addressing for communication
between these two processes still applies because their full address (service, logi-

er
cal, and device addresses) is used. To identify common network processes, a well known addresses:
group of service addresses, called well-known addresses, are set aside. Reserved service addresses
for common network
processes.

at
Well-known addresses
include the FTP port 21 and

y
HTTP port 80.

op
eM
tC
No
c
en
Do

Figure 7-1: How addresses work together.

se
r

How Addresses Work Together

fe
ea

• The Logical Network address identifies the source or destination network.

• The Physical Device (MAC) address identifies the source or destination
computer.
Pl

•
Re

The service address identifies the specific process that is running on the
source or destination computer.

Lesson 7: The Network Layer 189

 TASK 7A-1:
Examining Network Layer Addressing
Objective: Identify the term being described.

ial
1. Address of the cable only.

2. Combination of service, logical, device.

er
3. Address of the port or socket.

4. Device processes.

at
y
5. Address of the device only.

op
eM
CHECK YOUR SKILLS 7-1
Suggested time:
2 minutes
Why are bridges faster than routers? tC
No
Based on the structure of addresses, explain why bridges are typically faster
c

than routers.
en
Do

TOPIC 7B
se

Internetwork Connection Devices

r
fe

Just like network connection devices physically join segments of a network

together, internetwork connection devices physically join different networks
ea

together. To do so, they use specific hardware designed to connect independent

networks. By connecting networks, users can have access to resources that might
only be available on another network.
Pl
Re

There are three types of internetwork connection devices: routers, brouters, and
CSU/DSU. Although a network is joined to another, it maintains its own charac-
teristics and unique address.

190
Router
Routers, operating in the Network layer, use network address information to
move data through the best path to its destination. While bridges, hubs, and
repeaters connect segments of the same network, routers connect different
networks. Each network must have its own unique address. To connect different A router determines how data
networks, routers must be aware of network addresses. Because routers can deter- should be routed based on

ial
the network addresses in the
mine network addresses, they can be ″intelligent″ about routing data through a
packets it receives.
complex internetwork involving multiple networks.

Subnet router:
A device that connects

er
The independent networks that make up an internetwork are often referred to as networks and can determine
subnetworks, or subnets. A subnet is an independent network physically con- the best path for data when
nected to and part of an internetwork. multiple paths are available.

It is important for each subnet to have a unique logical address so that the router

at
can move data to the appropriate subnet. A logical address is assigned by an
administrator and can be modified at any time, unlike a physical address, which subnet:
is typically assigned by the hardware manufacturer and cannot be changed. Both An independent network
physically connected to and

y
addresses are unique; they help identify a single computer.
part of an internetwork.

op
eM
tC
No
c
en
Do
se
r
fe
ea

Figure 7-2: Routers function as traffic directors between networks.

Pl
Re

Lesson 7: The Network Layer 191

 A router might be a device designed specifically for routing, or it might be a
computer configured (with network adapters and routing software) to serve as a
router. When a file server acts as a router, the file server is said to have an inter-
nal router; routers that aren’t part of a file server are often called external
routers.

internal router: A router functions much like a bridge; however, there are important differences

ial
A file server that functions as aside from whether you transmit data between segments or networks. A router is
a router. more ″intelligent″ than a bridge and it requires more time to determine a data
path, so its processing speed (the time it takes to forward data blocks, or packets)
is typically slower than a bridge’s. However, a bridge cannot determine the most
efficient data path like a router can. The decision to use one or the other should
external router: be based on network needs and environment.

er
A router that isn’t part of a
file server.

Brouter
Because routers are dependent on Network layer communication protocols, they

at
can connect only networks that use communication protocols with which they are
familiar. If a router discovers that a data transmission is using an unfamiliar pro-
tocol, it simply drops the transmission.

y
Brouters are a hybrid of routers and bridges. Brouters function like routers, relay-
ing data transmissions between networks, but when they encounter a data unit

op
eM
that uses a protocol with which they are unfamiliar, they work like a bridge and
brouter: forward the data to the next segment by using a physical address. Because of this
A communication device, ability, you can use brouters for networks on which there is mixed-protocol traffic
often referred to as an and for networks that use protocols that do not support routing, such as NetBEUI

tC
internetworking device, that
functions as both a bridge
and DECnet LAT.
and a router. An advantage of brouters is that they combine the data-handling capabilities of
routers with the versatility of bridges. A disadvantage is that they are generally
No
more complex and expensive than other internetworking devices.
c

Channel Service Unit/Digital Service Unit

en
Do

Because the installation and upkeep of large quantities of transmission media can
become very costly, you might choose to use a public or private service organiza-
tion for your transmission media needs. By using one of these organizations, you
can connect to their media and save yourself the installation and maintenance of
your own.
se
r

Sometimes public service organizations require their customers to use channel

service units (CSU) and digital service units (DSU) in the interest of safety. The
fe

units are designed to shield network users from electric voltages as well as elec-
trical interference. They also ready data for transmission by adhering to any
ea

CSU/DSU: network rules.

(Channel Service Unit/Data
Service Unit) Combines the Both units are parts of a data communications equipment (DCE) device and are
functions of CSU and DSU collectively referred to as a CSU/DSU. A CSU/DSU operates like a modem; how-
Pl
Re

in one device. The CSU ever, it is a digital-to-digital device rather than digital-to-analog. It readies digital
terminates a digital circuit. signals and guarantees that they have sufficient strength and the proper format to
The DSU terminates a data be transmitted over digital WAN links.
circuit to the Data Terminal
Equipment and converts Typically, a CSU/DSU connects to a router or remote bridge by using a V.35 con-
customer transmission data nection or similar synchronous serial interface. It is also known as an integrated
into a bipolar format.
service unit (ISU).

192
TASK 7B-1:
Applying internetworking devices
Objective: Identify which network device would provide the necessary
services for the following services.

ial
1. Stone Investments needs to provide communication between different
networks. Most of the networks use the same protocols; however, some
use different ones.

er
2. Global Travel needs to connect several networks that use the same
upper-layer protocols. They want to use devices that will provide the
most efficient path among the networks.

at
y
3. Andy’s Supply House needs a high-speed WAN connection but can’t
afford the high cost of installation and maintenance that is associated

op
with owning the communication lines.
eM
TOPIC 7C tC
No
Routers
c

Devices on different networks communicate by using routers. Routers, together

with routing protocols, transfer data packets across internetworks. To do so, the
en

router must know the receiving device’s address and a path to the device. To find
Do

out where the receiving device lives, the router uses the device’s software
address, which distinguishes and describes the device and the network on which
it resides. (A device’s physical address describes only the device.)
A software address can be either logical or logical and physical.
se

•
r

Logical —requires mapping to a physical address before the packet can be

transmitted to the receiver.
fe

• Logical and physical —includes both the network address and a copy of the
ea

physical address.

Packet Routing
Pl
Re

In Figure 7-3, Node A resides on Network 2 and wants to transmit data to Node
B on Network 4. Routers 2 and 3 both use the same Network layer protocol. Net-
works 2 and 4 are Ethernet networks; Network 3 is Token Ring.

Lesson 7: The Network Layer 193

 ial
er
at
y
op
eM
Figure 7-3: Routing packets by using Network layer protocols and software addresses.
Node A’s upper-layer processes use the target software address to identify where

tC
the data is going, and then let Router 2’s protocol take over. The protocol sees
the address (Network4NodeB) and knows that the data is traveling to Node B on
Network 4. The protocol constructs a data packet and appends a Network layer
header of the target and source software addresses (Network4NodeB and
No
Network2NodeA, respectively). Realizing that Node B does not reside on Net-
c

work 2, Router 2 enlists the help of the Data Link layer to forward the packet.
The Data Link layer places the data packet in a MAC frame and adds a destina-
tion hardware address for Router 2 since the packet must pass through Router 2.
en

Upon receiving the packet, Router 2’s Data Link layer verifies the target software
Do

address and CRC, deletes the header and trailer, and then gives the rest to the
Network layer.
Router 2’s protocol looks at the packet’s target network section
(Network4NodeB) and sees that the packet must be delivered to a different net-
se

work (Network 4). It consults its route table to find a suitable route for the packet
r

to take to Network 4. Router 2’s route table indicates that Network 4 is only one
hop away and that Router 3 is directly linked to Network 4. Router 1’s protocol
fe

sends the packet back to the Data Link layer, instructing the layer to transmit the
ea

packet to Router 3.
Router 2’s Data Link layer then creates a Token Ring frame (because Network 3
is a Token Ring network) and sends it to Network 3. From there, Router 3’s pro-
Pl
Re

tocol receives the packet and verifies the network information and software
address. Realizing that the packet’s target node resides on Network 4, Router 3’s
protocol checks the packet’s software address for the target node to find out
exactly where the packet must go. The protocol then attaches instructions to for-
ward the packet to Node B and passes it back to the Data Link layer for delivery.

194
Upon receiving the packet, Node B checks the target software address and sees
that it is the intended recipient. It deletes the Network layer header and forwards
the data to its upper layers.

Route Tables

ial
A route table is a database created by a route discovery protocol and containing
network locations as perceived by a specific router. A router uses its route table to
forward packets to another network or router. The action of forwarding a packet
from one router to the next is called a hop. A router sends a packet to another Each router has a table from
router until the packet reaches the network on which its receiver is located. You its point of view.
can specify the number of hops that packets can take from sender to receiver.

er
The number of hops along a route between two networks constitutes that route’s
cost. However, a cost can also consist of other specifications, such as speed. Typi-
cally, a router maintains the most cost-effective route in its table.

at
y
op
eM
tC
No
Figure 7-4: A sample network with three routers: A, B, and C.
c

In the sample network shown in Figure 7-4, the routing tables for Routers A and
B would contain the following information:
en

Routing Table for Router A

Do

Destination Next Hop Type Cost Intf

140.240.0.0 140.240.20.254 Direct 1 1
192.233.66.0 192.233.66.253 Direct 1 2
136.54.0.0 192.233.66.254 Remote 2 2
se
r

126.0.0.0 192.233.66.254 Remote 3 2

fe
ea

Routing Table for Router B

Destination Next Hop Type Cost Intf
192.233.66.0 192.233.66.254 Direct 1 1
Pl
Re

140.240.0.0 192.233.66.253 Remote 2 1

136.54.0.0 136.54.20.254 Direct 1 2
126.0.0.0 136.54.15.254 Remote 2 2

Lesson 7: The Network Layer 195

 Consider network 140.240.0.0 to be network 1; 192.233.66.0 to be network 2;
136.54.0.0 to be network 3; and 126.0.0.0 to be network 4. Notice that each of
the tables above see the network from a different perspective.

Routing Protocols

ial
A routing protocol provides a router with capabilities, such as discovering paths
and services, failure notification, broadcasting route costs, and maintaining route
tables. The protocol resides in a Network layer packet and has no interaction with
Physical or Data Link layer processes. Its only concern is how the networks it is
servicing are connected.

er
Two algorithms are used to describe how a routing protocol creates and maintains
a route table and the method behind selecting a path to a network: distance vector
and link state.

at
Unroutable Protocols
Sometimes a protocol that cannot freely select a route over which to send packets

y
is referred to as unroutable or nonroutable. It is forced to route them by using
only static paths specified in a simple lookup table. This table cannot be modified

op
NetBIOS and LAT are
examples of protocols that
eM
or updated. A protocol that does not use the Network layer to route packets can
also be labeled unroutable.
were originally designed only
with LANs in mind. If you’re configuring a WAN, you can use routable protocols such as TCP/IP or

tC
IPX/SPX. Protocols such as DLC or NetBEUI, which are nonroutable, couldn’t
be used on a WAN.
Bridging and tunneling (encapsulation) are used in conjunction with unroutable
protocols. The following table lists some of these protocols.
No
c

Protocol Is unroutable because

NetBIOS:
Network Basic Input/Output NetBIOS It does not have a Network layer. It communicates by using a Session layer
System. protocol.
en

NetBEUI It is an extension of NetBIOS that is designed to carry data within only small- or
Do

medium-sized LANs.
SNA All of its routes are pre-defined.
NetBEUI: LAT It does not have a pure Network layer. It was designed to cross only a single
NetBIOS Extended User
network and lacks a header field for the logical network address.
Interface.
se
r
fe

SNA: Multi-purpose Devices

ea

Simple Network Architecture.

Networks often incorporate more than one technology in their structures and need
the capabilities of different connection devices. Two devices that combine func-
tions of different devices are a brouter and a switching hub and router.
Pl
Re

LAT:
Local Area Transport.

196
Brouter
A brouter does the jobs of both a router and a bridge but is contained in one
device unit. It is ideal for a situation in which you need to route and bridge mul-
tiple unroutable protocols. However, it is not cost-efficient to use a brouter if you
need only routing or only bridging capabilities because a brouter costs more than
a single router or single bridge.

ial
Switching Hub and Router
As the name implies, a switching hub and router combines the capabilities of a
switch and a router into one device unit. However, switching hubs and routers are

er
based on relatively new technology so there are not many products from which to
choose and those available can cause network problems.

TASK 7C-1:

at
Discussing unroutable protocols

y
1. How are routes selected?

op
eM
tC
2. What constitutes an unroutable protocol? List some examples of
unroutable protocols.
No
c

3. What type of connection device should you use for routing unroutable
protocols?
en
Do

Route Selection Process

se

Once a router’s table has been created, it’s ready for service. By using informa-
r

tion in its table, a router performs an algorithm to calculate the best possible path
for each piece of data it receives. However, the route selection process can be
fe

static or dynamic.
ea
Pl
Re

Lesson 7: The Network Layer 197

 Static Routing
Static selection means the path has been determined—either by a network device,
which places its own header information on the frame, thereby forcing a specific
path, or by a system administrator through his or her own creation of a router
table—the intermediary routers don’t have the authority to make any routing
decisions. Although static route selection is relatively inexpensive, there is little

ial
or no allowance for router failure or overworked routers. Thus, static route selec-
tion can be management-intensive.

Dynamic Routing

er
Dynamic selection uses the router’s algorithms and cost calculations to perform
the latest routing discovery for every packet. Data stops at each router in its path;
that router is responsible for determining the next best step. Dynamic route selec-
tion is highly flexible and responds well to changes in the routing environment
with little or no management. The main disadvantage is the need for complex and

at
relatively expensive routing devices.

y
Comparing Static and Dynamic Routing

op
eM
In summary, static routing is manually configured, and uses a prescribed route
that might or might not be the most efficient route. This method is more secure
than dynamic routing since the route is predictable.

tC
Dynamic routing is automatically configured (except you must manually enter the
first entry) and uses route metrics to determine the most efficient route. This
method is less secure than static routing due to possibly insecure routing paths
being dynamically selected.
No
Dynamic routing tables require less maintenance than static routing tables. How-
c

ever, static routing tables enable the network administrator to have more control
over the contents of the routing tables.
en
Do

TASK 7C-2:
Examining static and dynamic routing
se
r

1. What does static selection mean?

fe
ea
Pl
Re

198
2. What does dynamic selection mean?

3. Compare static and dynamic routing.

ial
er
Default Gateways and Subnetworks
Every network that is part of an internetwork requires a unique network ID so
that each network can be identified. Routers on the internetwork use these IDs to

at
determine where to send messages. All of the hosts within a network use the
same network ID. A router can be specified as the default gateway when config-
uring TCP/IP so that network hosts can reach external network hosts.

y
Each subnetwork is connected to the larger internetwork via a router (or other

op
eM
routable connection device). In order for the hosts on each subnet to be able to
connect to hosts on another subnet, a default gateway must be specified on each
host. This allows the host to reach a router that can then send the data on to
another router or to its final destination.

tC
default gateway:
Subnetworks are most frequently found on TCP/IP networks. Without a default An IP router that provides
gateway you won’t be able to communicate with other networks. Without an access to a remote network.
appropriate subnet mask (an IP address uses a subnet mask to divide the network
into several smaller subnetworks), you won’t communicate with anyone—not
No
even someone on your own local network.
c

Clients need both a subnet mask and a default gateway specified in their TCP/IP
configuration. The default gateway is the IP address of the router on their local
en

network that allows clients to reach beyond their local network.

Do
se
r
fe
ea
Pl
Re

Figure 7-5: How LANs connect using a default gateway to reach external networks.

Lesson 7: The Network Layer 199

 On an IP network, to configure the default gateway, specify the IP address of the
network device that connects the LAN to the external network.
In Figure 7-5, there are two LANs connected via routers using a Frame Relay
connection. The networks are configured using TCP/IP. The default gateway for
LAN A would be the address on the network side of its router, 192.178.200.1; the
default gateway for LAN B would be 192.178.201.1.

ial
Let’s take a look at how to configure these on your workstation.

TASK 7C-3:

er
Examining default gateway and subnets
1. Open the Network Control Panel on your workstation.

2. Display the TCP/IP properties.

at
3. Display the Gateway tab. What is the current gateway? If there isn’t
one listed, why not?

y
op
eM
4. Display the IP Address tab. Is there a subnet mask? Why or why not?

5. tC
Close the TCP/IP Properties window without making any changes.
No
6. Close the Network Control Panel.
c

Suggested time:
APPLY YOUR KNOWLEDGE 7-1
en
Do

5 minutes
Reviewing the Network layer
In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.

1. Load the Skill Assessment component of the Network+ Certification

se
r

CD-ROM. (Remember to accept the disclaimer.)

fe

2. Start the Network+ test.

ea

3. Select all items for the Network Layer topic, uncheck the Test Time check
box, and then click OK.
Pl
Re

4. Work through the sample questions and score the test.

5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

6. Exit the test.

200
 Summary
In this lesson, you’ve examined Network layer concepts, identified devices
working at this layer of the OSI model, and examined routable and
unroutable protocols.

ial
LESSON 7 REVIEW

er
7A Describe addressing associated with the Network layer of the OSI model.

at
7B Explain the purpose of network connection devices and describe issues
associated with them.

y
op
eM
tC
7C What are some of the unroutable protocols? No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 7: The Network Layer 201

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
The Transport Layer LESSON
8

ial
Overview Data Files:
The Transport layer is the 4th layer of the OSI Model. It is responsible for Network+ Certification
making sure a packet is delivered to its destination. It performs connection CD-ROM

er
services for the Network layer. It also provides communication services for
the layers above this layer. Lesson Time:
45 minutes

Objectives

at
To understand the functions of the Transport layer, you’ll:

y
8A Identify Transport layer addressing issues.

op
eM
One of the addressing issues you’ll examine is the purpose of name
resolution.

8B Describe the Transport layer segmenting process.

tC
In this topic, you’ll see how the Transport layer divides messages into
segments.

8C Define the three Transport layer connection services.

No
This topic covers connectionless and connection oriented services.
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 8: The Transport Layer 203

 TOPIC 8A
Transport Layer Addressing Issues
The Transport layer is responsible for reliable communication between the lower

ial
layers and upper layers. This layer organizes data into segments for the lower
layers and then combines the segments into messages for the upper layers. The
upper layers include service providers, such as file, print, and application servers.
The Transport layer:
• Takes action to correct faulty transmission.

er
• Controls the flow of data.
• Acknowledges successful receipt of data.
• Fragments and reassembles data.

at
Some examples of Transport
protocols include NETBEUI, Addressing Methods
XNS, SPX, AppleTalk

y
Transaction Protocol, The Transport layer is often thought of as the heart of the OSI model because it
Transmission Control provides upper layers with a communication channel. To do this, it uses services

op
Protocol (TCP), User
Datagram Protocol (UDP),
eM
provided by lower layers. Through addressing, successful communication employs
conversations between service requesters and clients. Each process running on a
and Windows NT’s Transport network device is identified by a service address specific to the Transport layer.
Driver Interface. The two addressing methods are:

tC
• Connection ID (connection identifier, or socket, or port).
• Transaction identifier.
No
connection identifier:
Connection Identifier
c

Also known as connection ID

socket or port. An address
that identifies each line of Typically, a service provider can converse with more than one client at a time.
The connection identifier or connection ID is an address that identifies each line
en

conversation to the upper

of conversation to the upper layers. The service provider communicates by this
Do

layers.
number, relying on the Transport layer to communicate each address with the
lower layers.

Transaction Identifier
se
r

The transaction identifier or transaction ID is much like the connection ID; how-
fe

ever, it works with smaller pieces of information rather than a whole

ea

conversation. A transaction consists of a request with a response; a conversation

transaction identifier: consists of one or more transactions. Transaction IDs enable the service provider
Enables the service provider to handle the necessary tasks for a request.
to handle the necessary
Pl
Re

tasks for a request. Similar

to connection ID, but works
on transactions rather than a
whole conversation. Also
known as transaction ID.

204
Address and Name Resolution
Computers are designed to function with the use of alphanumeric characters.
Humans, on the other hand, deal better with names. To bridge the gap between
the complicated network addresses, networking protocols offer a service called
address-to-name resolution. Address-to-name resolution is the process of mapping
an alphanumeric address to an understandable name. This service can be per-

ial
formed on every machine in a network or on a specialized service provider
known as a directory or name server. There are two methods for resolving names: address-to-name
the service-requester initiated method; or service-provider initiated method. resolution:
The process of mapping an
• In the service-requester initiated method, the requester submits a packet ask- alphanumeric address to an
ing for a response from the device that matches the submitted name, address, understandable name.

er
or service type. The matching device then responds with either its name or
address.
• In the service-provider initiated method, all service providers send a broad-
cast packet that includes the packet’s name and address. The network
devices or the specialized directory servers gather these broadcasts. With the

at
name and address information saved in a table, the directory server is able to
resolve names and addresses.

y
WINS

op
eM
The name you assign to a computer when you install Windows NT is the comput-
er’s NetBIOS name. For users, NetBIOS names provide an easy-to-use method to
access the resources of computers on the network. Any NetBIOS-based network

tC
utility can connect to a computer by using the computer’s NetBIOS name. How- WINS:
ever, most TCP/IP commands are designed to use TCP/IP addresses, not names. Windows Internet Name
The Windows Internet Name Service (WINS) automatically maps names for IP Service. Automatically maps
NetBIOS (computer) names
addresses in commands and utilities. (Any computer on a TCP/IP network that for IP addresses in
No
performs NetBIOS name resolution is called a NetBIOS Name Server (NBNS). commands and utilities.
c

WINS is one example of an NBNS.)

There are several reasons why you might implement WINS:
•
en

If you have a Windows NT network that is confined to one network seg- NetBIOS Name Server
Do

ment, TCP/IP broadcasts will resolve NetBIOS names to IP addresses. You (NBNS):
might choose to implement WINS in this situation if you want to reduce Any computer on a TCP/IP
broadcast traffic. WINS uses directed datagrams between the WINS server network that performs
and the other hosts to perform the name resolution. NetBIOS name resolution.
WINS is an example of a
• When you have UNIX hosts on the same network with Windows NT com- NBNS.
se
r

puters, you can enter static address mappings for those hosts’ names on the
WINS server. This enables Windows NT users to use commands like ″ping
fe

unix_host″ to communicate with the UNIX systems.

ea

• WINS can greatly simply the administrative burden of name resolution in an

internetworked environment. Without WINS, network administrators need to
maintain manual name resolution tables in LMHOSTS files on all the routers
in the internetwork.
Pl
Re

The WINS Process

The WINS process works as follows:

Lesson 8: The Transport Layer 205

 1. When a WINS client boots up, it sends its IP address and NetBIOS informa-
tion in a directed datagram to a WINS server. Because the WINS client uses
a directed datagram and not a broadcast, the WINS Server can be on a dif-
The WINS database contains ferent network segment from the client.
multiple entries for each
node, because different 2. The WINS Server enters this information in a dynamic database.
services on each computer 3. When a WINS client needs to connect to another computer, the client will

ial
register separately.
query the WINS Server directly to get the IP address for that computer,
instead of broadcasting name queries.
4. The client receives the requested NetBIOS name to IP address mapping from
the WINS Server and stores the information in its NetBIOS name cache.
Thus, the NetBIOS name for each target computer may need to be resolved

er
to its IP address only once via WINS, unless the client computer’s NetBIOS
name cache entry is purged or expires.

TASK 8A-1:

at
Addressing and the Transport layer

y
Objective: To identify addressing issues in the Transport layer.

op
1.
eM
What are two other names for ″socket″?

tC
2. A transaction consists of what two things?

3. List five service providers.

No
c

4. What two addressing methods are in the Transport layer?

en
Do

5. Define connection ID in your own words.

se
r
fe
ea
Pl
Re

206
TOPIC 8B
The Transport Layer Segmenting Process
The Transport layer is responsible for dividing large upper-layer messages into

ial
easier-to-use segments for the lower layers. Conversely, the Transport layer can
also combine many smaller messages into segments for efficient communication
and minimized overhead. To combine smaller messages into a segment, all mes-
sages must have the same destination address.
Messages are made up of the following components: a logical network address, a
physical device address, a service address, connection identifiers, and data, as in

er
Figure 8-1. The components appear in the order that they are listed from left to
right. When smaller messages are combined, their unique connection identifiers
separate their data, as in Figure 8-2.

at
y
op
eM
tC
No
c
en
Do
se
r

Figure 8-1: Breaking large messages into segments for transport.

fe
ea
Pl
Re

Lesson 8: The Transport Layer 207

 ial
er
at
Figure 8-2: Combining small messages for transport.

y
op
eM
Segment Sequencing
When a message is divided into segments to prepare for departure on a packet-
switched network, it undergoes a synchronization process. During the segment

tC
sequencing process, a number is placed on each segment for proper sequencing of
all pieces when they arrive at their destination. At the destination, the process of
segment synchronization is used to organize and return the segments to their
original order.
No
Two factors can make this task a complicated one. First, the amount of memory
c

allotted to the Transport layer might not be adequate. Also, there is a chance of
losing segments. To help decrease the potential risks, connection services are also
performed in the Transport layer.
en
Do

TASK 8B-1:
Reviewing the segmenting process
se
r

Objective: Identify whether each of these statements are true or false.

fe

1. Segments are divided into messages.

ea

2. Only one message can be sent out at a time.

Pl
Re

208
3. A practical function of the Transport Layer segment process is to help
reduce overhead.

4. Segment sequencing assigns sequential numbers to segments before they

are transmitted. Upon arrival at the destination, the out-of-order seg-

ial
ments are returned to their original order.

TOPIC 8C

er
Transport Layer Connection Services
Like the Network and Data Link layers, the Transport layer also provides differ-
ent types of connection services. The Network layer deals with traffic between

at
intermediate nodes and the Data Link layer handles device-to-device traffic. The
Transport layer is concerned with the conversation between two end node
connections.

y
To guarantee reliable end-to-end communication, an acknowledgment might be

op
eM
needed at the Transport layer even though one occurred at the Network layer. The
Transport layer tries to ensure the reliability of data delivery by detecting and
attempting to correct problems that might have occurred in the lower layers (such
as missing or improperly sequenced packets of data). If the Transport layer can-

tC
not actually correct data-transmission problems, it sends a message to the upper
layers that a problem has occurred; then it can request re-transmission.
The specific connection services implemented in the Transport layer are:
•
No
Unacknowledged connectionless service —does not acknowledge data that
c

has been successfully transferred; layers above must provide their own reli-
ability checks (flow, error, and packet sequence).
• Connection-oriented service —establishes and maintains a virtual connection
en

between two nodes. This service performs flow control, packet sequencing,
Do

and error-recovery functions through acknowledgment and reliable end-to-

end connections.
• Acknowledged connectionless service —combines features of the other two
services. It is similar to unacknowledged connectionless service, except that
it provides a mechanism through which individual packets can be
se
r

acknowledged.
fe
ea

End-to-end Error Control

To eliminate or recover from errors-such as segments arriving with duplicate seg-
ment numbers or segments not arriving at all, the Transport layer can:
Pl
Re

• Make the duplicate number unique.

• Use only virtual circuits instead of packet-switching communication services.
• Limit virtual circuits to one per session.
• Use a time-out process to drop packets that have been circulating too long.

Lesson 8: The Transport Layer 209

 The Transport layer can perform Cyclic Redundancy Checks (CRCs) or
checksums to verify that data and header information has not been corrupted.

End-to-end Flow Control

ial
To control the flow of data, the Transport layer uses acknowledgments and either
a guaranteed rate method or a window flow control method to request retransmis-
sion of segments that were not received. There are two types of acknowledgments
The two types of that the Transport layer uses: Go Back N and Selectively Repeat.
acknowledgements used in
the Transport layer include The Go Back N acknowledgment instructs the sender to re-send parts of the mes-
Go Back N, which instructs sage beginning with the packet immediately following the last successfully

er
the sender to re-send parts received packet. The Selectively Repeat acknowledgment instructs the sender to
of the message beginning resend only the packets that were not received, regardless of their location in the
with the packet immediately
following the last
string.
successfully received packet,
and Selectively Repeat,

at
which instructs the send to
resend only the unreceived
packets.
TASK 8C-1:
Discussing connection services

y
op
1.
eM
Define end-to-end flow control.

tC
2. List the four ways to control errors.
No
c

3. How do Go Back N and Selectively Repeat differ in strategy?

en
Do
se
r

Transport Layer Protocols

fe
ea

Transport layers protocols set up the communication session between computers

and ensure that data is able to move between computers reliably. Examples of
popular Transport layer protocols are listed in the following table.
Pl
Re

Transport Layer
Protocol Description
TCP (Transmission Control Protocol) Part of the TCP/IP protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.
SPX (Sequenced Packet Exchange) Part of Novell’s IPX/SPX protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.

210
Transport Layer
Protocol Description
NWLink Part of the Microsoft version of Novell’s IPX/SPX protocol suite.
Responsible for guaranteeing delivery of data packets in sequence.
NetBEUI (NetBIOS Extended User Interface) NetBIOS (Network Basic Input/Output
System) is responsible for establishing the communication session, and

ial
NetBEUI is responsible for providing the data-transport services.
ATP (AppleTalk Transaction Protocol) Responsible for establishing the
communication session and providing data-transport services in Apple
computers.
NBP (Name Binding Protocol) Responsible for establishing the communication
session and providing data-transport services in Apple computers.

er
Connectionless and Connection-oriented Protocols

at
Let’s examine the use of connectionless and connection-oriented protocols.
Connectionless protocols send datagrams (the term for packets in a connectionless

y
environment) across the network without any knowledge of whether they actually
get to their destination. This is a fast method of sending information since there

op
eM
isn’t any setup or tear-down of connections. A connectionless protocol doesn’t connectionless protocol:
A data communication
use guaranteed delivery service in the Transport layer; upper layers of the OSI
model perform this service. If you use a connectionless protocol and a packet method in which frames are
″blindly″ sent over media
gets lost, the Transport layer has no method for finding it—the protocol relies on

tC
without knowing if the
other layers of the OSI model to find it or have it retransmitted. Another name receiver can receive them.
for this type of connection is best-effort connections .
Examples of connectionless protocols include IP, IPX, and UDP protocols.
No
Connection-oriented protocols require that two (and only two) systems create a
c

connection before any data is transferred. The packets are then sent according to
the rules established by the protocol in use. This includes acknowledging receipt
of the packet and re-sending the packet if there is any error. This method is connection-oriented
en

slower than connectionless protocols, but is very reliable. protocol:

A data communication
Do

Examples of connection oriented protocols include TCP and SPX. Also, modem method in which packets are
connections between two systems would be connection oriented. sent and acknowledged by
the recipient and resent if
there is any error.

TASK 8C-2:
se
r

Connection-oriented or not?
fe
ea

Objective: Identify whether each of the following are connection-oriented

or connectionless protocols.
Pl

1. IP
Re

2. IPX

Lesson 8: The Transport Layer 211

 3. TCP

4. UDP

ial
5. SPX

APPLY YOUR KNOWLEDGE 8-1

er
Suggested time:
5 minutes
Reviewing the Network layer
In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.

at
1. Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)
2. Start the Network+ test.

y
3. Select all items for the Transport Layer topic, uncheck the Test Time check

op
eM
box, and then click OK.
4. Work through the sample questions and score the test.
5. If you want to review the questions and the correct answers, click Review

tC
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.
6. Exit the test.
No
c

Summary
en
Do

In this lesson, you’ve examined Transport layer addressing, name resolution,

segmentation of messages, and protocols at the Transport layer.

LESSON 8 REVIEW
se
r
fe

8A Describe the difference between the connection identifier and transaction

ea

identifier addressing methods.

Pl
Re

212
8B Describe the process of segmenting.

ial
8C Define the three connection services in the Transport layer.

er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 8: The Transport Layer 213

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
The Session Layer LESSON
9

ial
Overview Data Files:
This lesson covers the fifth layer of the OSI model. The Session layer is none
responsible for management of simplex, half-duplex, or full-duplex transmis-
Lesson Time:

er
sions between computers.
30 minutes

Objectives

at
To understand the functions of the Session layer, you will:

9A

y
Describe the steps for Dialog control.
Dialog control affects data flow between nodes in either simplex, full-

op
eM
duplex, or half-duplex as described in this topic.

9B List and define steps for managing the session.

tC
This topic discusses establishing the connection, transferring the data, and
releasing the connection.
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 9: The Session Layer 215

 TOPIC 9A
Dialog Control Modes
The Session layer establishes and maintains the connection between service pro-

ial
viders and clients that communicate through the network. To do this, it must
employ services provided by the Transport layer.
Examples of Session layer protocols include DNA Session Control, AppleTalk
Session Protocol, and NetWare Service Advertising Protocol (SAP).
The Session layer:

er
• Establishes and maintains connections.
• Manages upper-layer errors.
• Handles remote procedure calls.
• Synchronizes communicating nodes.

at
The Session layer comes into play primarily at the beginning and end of a
transmission. At the beginning of the transmission, it makes known its intent to

y
transmit. At the end of the transmission, it determines if the transmission was
successful.

op
eM
Network nodes can interact on a hierarchical, client-server, or peer-to-peer basis.
Typically, it is in the Session layer that the type of relationship between nodes
and the dialog control methods, which control data flow are established.

Dialog Control Methods

tC
Dialog control affects data flow between nodes in one of three ways: simplex,
No
full-duplex, and half-duplex.
c

Simplex Transmission
en

In a simplex transmission , only one device transmits at a time—all other devices

Do

receive. The communication channel is dedicated to one-way transmissions. Both

the transmitter and receiver are limited to a single function; transmitters cannot
simplex transmission: receive information and receivers cannot transmit. Examples of simplex transmis-
One-way transmission. sions are radio and television broadcasts.
Benefits of simplex transmission include an ability to cover a large area and to
se
r

target large audiences, inexpensive hardware, and a lack of channel contention.

The main disadvantage is that only one-way communication is supported.
fe
ea
Pl
Re

Simplex transmission is like

a one-way road. Travel is
only allowed in one
direction.

216
 ial
er
at
y
Figure 9-1: Radio broadcast is an example of simplex transmission.

op
eM
Full-duplex Transmission

tC
Full-duplex transmission uses two communication channels so that devices (usu-
ally two) on each end of a transmission can transmit and receive data at the same
time. One channel moves data in one direction, and the other channel moves data
in the other direction. Telephones use this mode of transmission; the mouthpiece full-duplex:
No
on each end transmits sound to the earpiece on the other end. Although it would Simultaneous two-way
c

make for a confusing conversation, both parties could talk at the same time if channel transmission.
they wanted to. Hardware for this transmission is more expensive than for either
simplex or half-duplex and requires more hardware and sometimes more
en

software. With full-duplex transmission, the audience is more exclusive.

Do

Full-duplex transmission is
like a two-lane road. Travel
se
r

is permitted in both
directions, each direction of
travel in a designated lane.
fe
ea
Pl
Re

Lesson 9: The Session Layer 217

 ial
er
at
Figure 9-2: Telephone conversation is an example of full-duplex transmission.

y
op
eM
Half-duplex Transmission
Half-duplex transmission is a compromise between simplex and full-duplex. In
half-duplex mode, there is only one channel. However, both devices can transmit

tC
and receive by sharing the channel. When one device transmits data, the other
half-duplex: device receives. Citizen’s band radio and most LAN data transmission use this
Data transmission on a mode.
channel that permits
No
communication one way at a The advantages of half-duplex transmission is that bi-directional communication
time. is possible and only one channel is required for transmitting and receiving. There
c

are, however, several disadvantages with this mode. While the communication is
bi-directional, only one device can transmit at a time. Also, more hardware is
required so it is more costly than simplex transmission. Finally, the channel is
en

Half-duplex transmission is inefficiently used because of the wait process involved while the setup for direc-
Do

similar to a one-lane bridge

on a two-lane road. Travel
tion change takes place.
over the bridge is permitted
in both directions, but not at
once. You must wait until
traffic traveling in the
se
r

opposite direction clears the

bridge before you can pass
over the bridge.
fe
ea
Pl
Re

218
 ial
er
at
y
Figure 9-3: Intercom system is an example of half-duplex transmission.

op
eM
TASK 9A-1:
Session layer dialog control modes
1. Provide three examples of half-duplex transmission.
tC
No
c
en
Do

2. Provide three examples of simplex transmission.

se
r

3. Provide three examples of full-duplex transmission.

fe
ea

4. What is the control mode used by most LANs?

Pl
Re

Lesson 9: The Session Layer 219

 TOPIC 9B
Managing the Session
The Session layer is responsible for managing the session, or communication,

ial
between interacting nodes. Managing the session involves managing three differ-
ent steps between the nodes. They are:
1. Establishing the connection.
2. Transferring the data.
3. Releasing the connection.

er
Establishing the Connection
For communication to occur, a connection must take place. Several criteria must

at
be met before any connection can take place. The first step in creating a connec-
tion is verification of login names with passwords. Once verification occurs,
connection ID numbers are established to ensure the agreed-upon services will be

y
associated and connected with each other, and the duration is specified for the
transmission. A negotiation must then transpire to determine the first speaker. This

op
eM
is followed by coordinating the Transport layer’s acknowledgment numbering and
procedures for re-transmission.

tC
Transferring the Data
This part includes the actual data transfer with whatever acknowledgment proce-
dure has previously been set. The transfer is either a continuous communication
No
or, if it happens to be interrupted, transfer occurs upon resumption of
c

communication.
en

Releasing the Connection

Do

A loss of connection creates a connection release, which takes place when the
agreed-upon acknowledgment procedure does not occur. A connection release also
occurs when both parties agree that the session is over. If desired, the session can
be rebuilt or a new session can be started.
se
r
fe
ea
Pl
Re

220
TASK 9B-1:
Managing the session
1. Acknowledgment of the reception of data is part of which managing
step?

ial
2. A connection release occurs in which two circumstances?

er
3. List the process for establishing a connection.

at
y
op
eM
Summary

tC
In this lesson, you’ve learned about full-duplexing, half-duplexing, and
simplex connections and how the Session layer manages communication
sessions.
No
c

LESSON 9 REVIEW
en
Do

9A List and describe the three dialog control modes.

se
r
fe
ea

9B List and briefly define the three steps for managing a session.
Pl
Re

Lesson 9: The Session Layer 221

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
TCP/IP Fundamentals LESSON
10

ial
Overview Data Files:
This lesson introduces TCP/IP fundamentals including IP default gateways, Network+ Certification
use of DHCP, DNS, WINS and Host files, the protocols that comprise the CD-ROM

er
TCP/IP Suite, the use of TCP/IP across all major operating systems through-
out the world, and how Internet domain name server hierarchies work. Lesson Time:
1 hour, 30 minutes

Objectives

at
To learn about TCP/IP fundamentals, you will:

y
10A Discuss the origins and purpose of TCP/IP.

op
eM
You’ll be examining where TCP/IP grew out of and compare the OSI and
DOD communication models.

10B Identify the protocols in the TCP/IP protocol suite.

tC
This topic discusses several of the protocols contained in the TCP/IP pro-
tocol suite and the purpose they serve.
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 223

 TOPIC 10A
What is TCP/IP?
Ideally, every computer would exchange information freely with every other com-

ial
puter, regardless of the name, type of processor, and operating system. In the real
world, however, even computers from the same manufacturer can have problems
exchanging information. One of the most complicated challenges that today’s sys-
tem integrators face is connecting different types of computers in a network.
TCP/IP, or Transmission Control Protocol/Internet Protocol, is a suite of proto-
cols that defines formats and rules for connecting networks of dissimilar

er
computers. TCP/IP is an internetworking solution that has become a standard for
providing internetwork connectivity.
TCP/IP refers to the suite, or collection, of protocols used by some computers to
communicate with one another. The two main protocols in the suite are:

at
• Transmission Control Protocol (TCP) —establishes inter-system communica-
tion
• Internet Protocol (IP) —transfers data

y
Be aware of the difference
between TCP/IP, which is the TCP/IP was developed by the Defense Advanced Research Projects Agency

op
entire suite of protocols, and
IP, which is a protocol within
eM
(DARPA) in the mid-1970s. Using TCP/IP, DARPA implemented an internetwork
the TCP/IP protocol suite. called ARPAnet. Today, the Internet and most modern NOSes have taken advan-
tage of the availability and standardization of TCP/IP software.

tC
TCP/IP is an industry standard, rather than a proprietary networking solution such
as NetWare, NT Server, or AppleTalk. Because it is an open standard, the specifi-
cations for the TCP/IP protocol suite are available to all application developers.
Also, many vendors carry products that support the TCP/IP protocol suite, so
No
TCP/IP can be used with virtually any type of computer and operating system.
For example, because most Web servers and browsers are designed to use TCP/IP
c

connections, you need only a browser and the TCP/IP protocol suite to access
Internet or intranet information from:
•
en

SUN, Apple, DEC, IBM, Intel-based clone, VAX, mainframe, mini, and
Do

other computers.

TCP/IP implementations that

• DOS, Macintosh, VMS, UNIX (4.2 and 4.2 BSD, and AT&T System V),
follow the specifications are Windows 95, Windows NT Workstation, and other operating systems.
compatible with one another,
no matter what platform
The TCP/IP suite orchestrates communications by transmitting information over a
network in packets. Every file transmitted on the network is divided into packets
se

they’re written to or what

r

vendor they come from. at the source node and reassembled at the destination node. TCP/IP protocols
define the structure of these packets, including the source address, the destination
fe

address, the length, and the type of packet.

ea
Pl
Re

224
 ial
er
Figure 10-1: The Internet.

The Internet

at
TCP/IP forms the base technology for the Internet: an internetwork that connects
research organizations, universities, corporations, government institutions, librar-

y
ies, and individual users. The following networks are part of the Internet.
The Internet evolved from

op
Network Description
eM ARPAnet, an experimental
network that connected
ARPAnet Defense Advanced Research Project Agency (DARPA, formally called ARPA). defense sites and
MILNET Department of Defense Network (originally part of the ARPAnet). universities, to a global

tC
internetwork that is
NSFnet A collection of networks funded by the National Science Foundation. constantly changing.
CSNET Computer Science NETwork (originally funded by the National Science Foundation).
Cypress Net Low-cost Internet access for small schools, individuals, and so forth.
No
X25Net Runs Internet protocols over Public Data Network (PDN).
c

Dial-up IP Part of CSNET developed for sites that use the Internet infrequently and, therefore,
cannot justify leased-line connections.
en

In addition to being the standard networking protocol for the Internet, TCP/IP is
Do

also the standard protocol for UNIX and Microsoft networks. And because
intranets are basically scaled-down versions of the Internet, TCP/IP is also the
standard networking protocol for intranets.
Internet services that the TCP/IP suite provides, such as HyperText Transfer Pro-
se

tocol (HTTP), Simple Mail Transport Protocol (SMTP), and File Transfer
r

Protocol (FTP), contribute greatly to its popularity. These are used for Web pages,
email and anonymous FTP services. Other services include USEnet news and
fe

gopher services.
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 225

 TASK 10A-1:
Discussing basic TCP/IP concepts
1. What does TCP/IP stand for?

ial
2. What is a major advantage of using the TCP/IP protocol suite?

3. How does TCP/IP transmit data over networks?

er
4. What types of general services are provided on the Internet?

at
y
The OSI and DOD Communication Models

op
eM
While the OSI model was defined by the International Standards Organization
(ISO) for computer networking, TCP/IP was developed by using a model devel-
oped by the Department of Defense (DOD). To better understand TCP/IP, it is

tC
important to understand how the DOD and OSI models compare.

The OSI Model

No
As you know, each of seven layers in the OSI model performs specific tasks that
c

are unique to that layer. Each layer directly communicates with adjacent layers,
and indirectly with its peer layer on another host.
en

Layer Description
Do

Application Specifies the interface with the user and manages the communication between
applications.
Presentation Formats the data so that it can be understood by each running application. It
also compresses, expands, encrypts, and decrypts data.
se

Session Initiates and terminates communications on the network. It also adds control
r

to the dialogue between communicating applications.

Transport Controls the quality and reliability of the data transmission. It also handles
fe

acknowledgment messages, sequence numbers, and flow control.

ea

Network Routes frames across an internetwork by examining the destination network

address and then sending the packet of information to the next node.
Data Link Organizes the Physical layer bits into logical groups of information called
Pl
Re

frames.
Physical Specifies the electrical and mechanical characteristics of the communication
medium and the interface device. The Physical layer accepts bits from a peer
Physical layer of another host and passes them on to the Data Link layer.

226
The DOD Model
The DOD model contains four layers representing the TCP/IP protocol suite.

Corresponding OSI
DOD layer Description layers

ial
Process/Application Provides the user interface and enables Application, Presentation,
users to run applications that access and Session
services on a TCP/IP network: file transfer
(FTP), terminal emulation (Telnet),
electronic mail (SMTP), and network
management (SNMP).

er
Host-to-Host Establishes and maintains concurrent Transport
connections between applications on
different hosts. This layer can provide
reliable transport (TCP) and unreliable
transport (UDP).
Internet Handles communication from one machine Network

at
to another and is therefore responsible for
routing packets optimally between different
hosts or different networks. This layer is

y
represented by the following protocols:
Internet Protocol (IP), Internet Control

op
eM
Message Protocol (ICMP), Address
Resolution Protocol (ARP), and Reverse
Address Resolution Protocol (RARP).
Network Access Receives and transmits IP datagrams Physical and Data Link

tC
between hosts, and is usually represented
by a device driver. The most common
protocols are Ethernet, Token Ring, and
Token Bus.
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 227

 ial
er
at
y
op
eM
tC
Figure 10-2: The DOD model, as compared to the OSI model.
The following steps describe how information flows in the DOD model.
1. At the Process/Application layer, an application such as FTP passes its data
No
to the Host-to-Host layer.
c

2. At the Host-to-Host layer, the TCP or UDP protocol adds to the data a
header that contains information about the application that originated the
data. A TCP header includes the source and destination ports that identify the
en
Do

upper-layer protocol; sequence and acknowledgment numbers; the size of the

header; flags to establish, control, or terminate a connection; the maximum
amount of data a host can accept (sliding window); and a checksum. This
header and the application data form a message that is sent to the Internet
layer.
se

3. The Internet layer adds the IP header to the message. This header includes
r

the software addresses of the source and destination, a reference to the Host-
to-Host layer protocol that should receive the message, flags to control the
fe

fragmentation and assembly of the packet, identification numbers for seg-

ea

mented packets, the total packet size, a checksum, and the packet’s Time To
Live (TTL). The IP header and the message form a packet that passes to the
Network Access layer.
Pl
Re

4. The Network Access layer adds a Media Access Control (MAC) header to
the packet. The MAC header contains the hardware addresses of the source
and the destination as well as a cyclic redundancy check (CRC) to ensure
data integrity. The MAC header, along with the packet, creates a frame.

228
5. When the frame arrives at a receiving end, data passes through the same
communication layers in reverse. At each layer, the corresponding header is
stripped off and processed. The remaining data is then passed to the next
layer until it finally reaches the application.

ial
TASK 10A-2:
Comparing the OSI and DOD communication models
1. Which DOD layer corresponds to the OSI Transport layer?

er
2. Which OSI layers correspond to the DOD Network Access layer?

at
3. Which DOD layer is responsible for routing packets between computers
or networks?

y
op
4.
eM
Which DOD layer enables users to run Internet applications such as
FTP?

TOPIC 10B tC
No
The TCP/IP Protocol Suite
c

Developed by the Department of Defense, the TCP/IP protocol suite is a standard

set of protocols for the communication and interconnection of all types of com-
en
Do

puter systems. The name is derived from the two most widely implemented
protocols in the suite: Transmission Control Protocol (TCP) and Internet Protocol
(IP). TCP and IP, however, are not the only protocols that form the suite.
se
r
fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 229

 ial
er
at
y
Figure 10-3: The TCP/IP protocol suite.

op
eM
RFCs

tC
The specifications of each TCP/IP protocol are maintained in the form of one or
more RFCs, or Requests for Comments. Each RFC is assigned a number and dis-
tributed over the Internet for comments and suggestions. When an RFC is
approved, it becomes an official protocol standard.
No
Be aware that there might be more than one RFC related to a topic. As new or
c

revised specifications are developed and submitted, they are given a new RFC
number.
en
Do
se
r

RFC:
Request For Comments. A
fe

numbered series of
documents containing
ea

information and standards

relating to the Internet.
Documents are posted, then
others can review and
Pl
Re

comment on them.

230
The Internet FAQ Consortium’s Archives Web site enables you to search for
RFCs by topic, so you can obtain all the RFCs that relate to the subject you seek.
In addition, the Internet Advisory Board periodically publishes an RFC, such as
RFC 2500, that lists the current official protocol standards for the Internet. You can obtain the text of
any RFC by visiting the
InterNIC Web site at http://
www.faqs.org/rfcs.

ial
Network Layer
The Network layer of the DOD model is equivalent to the Physical layer of the
OSI model. Specifications for boards, connectors, cabling, and network topology
are defined in this layer.

er
Internet Protocols
TCP/IP contains several protocols, such as IP. They can be categorized according
to the DOD model as Internet layer protocols, address resolution protocols, host-
to-host protocols, and process/application protocols.

at
Internet Layer Protocols

y
op
eM
Since the Internet is made up of networks connected by routers over various
types of WANs, data must sometimes travel through several routers before reach-
ing its destination. The Internet layer protocols ensure that data reaches its
destination. Its main job is to provide packet routing, fragmentation, and

tC
re-assembly between hosts.
The Internet layer protocols include IP, DNS, ICMP, ARP, RARP, and BootP.
No
Remember that the Internet
c

layer corresponds to the OSI

Network layer.
en
Do
se
r
fe
ea
Pl
Re

Figure 10-4: Internet layer protocols.

The following table describes some common Internet layer protocols.

Lesson 10: TCP/IP Fundamentals 231

 Name RFC Description
Internet Protocol (IP) 791, 950, 919, Defines the structure of a datagram, and sets the
922 rules for routing services and for packet
fragmentation and re-assembly.
Internet Control Message 792 Used by routers and hosts to broadcast control or
Protocol (ICMP) error messages. These messages inform other

ial
routers and hosts about optimal paths for moving
packets, whether a host can be reached, and so
on.
Domain Name System (DNS) 1034, 1035 Used for resolving a host name into the
corresponding IP address.

er
IP
IP is a connectionless, unreliable, packet-switching protocol. It enables packet

at
routing, fragmentation, and re-assembly of Transport layer packets, or Transport
Protocol Data Units (TPDU). Each TPDU is sent as a frame and has a header
containing information as displayed in the following figure.

y
op
eM
tC
No
c
en
Do
se
r
fe
ea

Figure 10-5: A TPDU IP header.

Pl
Re

IP carries the packets along the network and through various routers, if necessary,
to their final destinations. It forwards the data one hop at a time, leaving indi-
vidual routers to determine the packet’s next hop. In situations where data packet
fragments use different routes to get to their destination, IP reorders the packet
once all of its fragments arrive at the destination.

232
ICMP
ICMP is essentially an extension of IP. It allows internetwork error reports, test
packets, and other informational messages related to IP to be generated for TCP
as well as other Upper-layer Protocols (ULPs). IP hosts issue ICMP messages
under the following instances:
•

ial
To discover which network it is attached to.
• To signal broken or congested links to the internetwork.
• A packet cannot get to its receiver.
• An IP header is invalid or experiencing problems.
• The buffer is full or too small to store a packet.

er
• A packet reaches its specified Time To Live (TTL).
• Upon discovering a faster route to a destination than what it is supposed to
use.
• To check if another host is online.

at
y
op
eM
tC
No
c

Figure 10-6: ICMP messages.

The following table includes typical ICMP messages.
en
Do

ICMP Message Description

Redirect Sent to the transmitting host by an intermediate router telling it
about a more efficient path to the receiver. ICMP can automatically
add paths to an IP host’s route table even if link-state routing is
se

not enabled on the host.

r

Source Quench Tells the sender to slow down (quenches) the rate at which it
sends data because a device (receiver or intermediate router) on
fe

the route cannot keep up.

ea

Echo Request/Echo Reply Determines whether two network devices can communicate. The
sender issues an Echo Request; the receiver responds with an
Echo Reply. The Ping utility employs this technology.
Pl

Time Exceeded Issued when a packet’s TTL expires.

Re

Destination Unreachable Issued when transmission problems prevent a packet from being
delivered to its destination.

ICMP requires no configuration and is supported by all TCP/IP implementations.

Lesson 10: TCP/IP Fundamentals 233

 ICMP Router Discovery
The ICMP implementation used in NT and NetWare 5.0 networks supports router
advertisement and soliciting. When you start a TCP/IP host, it can send a
multicast packet requesting identification from neighboring routers. Routers sup-
ICMP router discovery can porting ICPM router discovery (RFC 1256) respond to the multicast by sending
also be configured to use the their addresses to the requesting host.

ial
broadcast address 255.255.
255.255.

er
at
y
op
eM
tC
No
c

Figure 10-7: ICMP router discovery.

Although ICMP router discovery gives a host neighboring router addresses, it is
en

not a routing protocol because it does not provide any information regarding
Do

paths to use to reach a given destination address. A host that selects an inefficient
route receives an ICMP Redirect message from an intermediate router, giving the
host a shorter route to the destination.
A host must be configured with a default route before it can receive ICMP
se
r

redirects. You can use ICMP router discovery as an alternative to configuring a

default static route.
fe

Important note: Do not enable ICMP router discovery if the routers on a network
ea

segment do not support router discovery.

Pl
Re

234
Domain Name System (DNS)
Domain Name System (DNS) is a shared database system that assigns or translates
host names to IP addresses for computers on an internetwork, and, more specifi-
cally, the Internet. To connect to a site (and thus a server), you can type an IP
address into a browser’s Location field, but since strings of numbers are harder to
remember than names, names are usually used. Also, the name stays the same if

ial
the IP address changes, so the name will always take you to the correct site.
DNS is a two-part service consisting of a hierarchy and a name service.
• The DNS hierarchy defines the structure, the naming conventions, and the
distribution of authority.
• The DNS name service maps the host names to the IP addresses.

er
For larger networks, DNS servers keep track of all host names and IP addresses.
You can get DNS services from your ISP, or you can use the DNS server soft-
ware provided with your operating system.

at
y
op
eM
tC
No
c
en

Figure 10-8: A DNS domain.

Do

DNS Hierarchy
DNS uses a hierarchical tree structure where the root is at the top. The DNS hier-
se
r

archy is also called the domain name space. Every node on the tree represents a
domain. The structure begins at the root and is followed by the top-level domains
that are divided into categories, such as .com and .org. The top-level domains are
fe

logical divisions, which can encompass many networks and domains. The top-
ea

level domains contain domains that represent organizations that can be further
divided into subdomains. (Domains nested in other domains are sometimes
referred to as subdomains. ) Typically, subdomains are further divided and branch
Pl

out to include more domains and subdomains. Domains at the end of branches
Re

represent a physical, addressable Internet computer.

Lesson 10: TCP/IP Fundamentals 235

 ial
er
at
y
op
eM
Figure 10-9: Domain name space.

tC
Top-level domain names include:

Internet domains Used by

INT (International) International organizations and organizations created by international
No
treaties.
c

GOV (Government) U.S. government agencies.

MIL (Military) The U.S. military.
COM (Commercial) Corporations.
en

EDU (Educational) Educational institutions.

Do

NET (Network) Internet Service Providers (ISPs).

ORG (Organizations) Miscellaneous other organizations.
XX Countries. Two-letter codes for each country. For example, NZ (New
Zealand) and GR (Greece).
se
r

Following are some examples of Country code domain names.

fe
ea

Two-letter code Country

AU Australia
BI Burundi
Pl
Re

BR Brazil
CA Canada
DE Germany
GR Greece
IE Ireland
JP Japan

236
Two-letter code Country
KR Korea
NZ New Zealand
UK United Kingdom
US United States

ial
Your organization might be divided into multiple subdomains. These domains
might be logical, as they might not follow the physical delineation of your
organization.
For a complete list of
country codes and manager

er
contacts, visit http://www.dir.
DNS Names org/country.html.

DNS names must be unique. A DNS domain’s complete name lists the path from
the domain to the root. A period is used to separate the labels (names in the
path); however, there is no period at the beginning of the name. There is a period

at
at the end of the name, but people often omit it. Domain names are not case-
sensitive and can be as long as 255 characters (however, sometimes a filename at
the end of the domain name is case-sensitive). Each label cannot exceed 63

y
characters. An example of the absolute or fully qualified domain name for the
ROCH domain in Figure 10-9 is:

op
ROCH.GO.COM.
eM
tC
Acquiring a Domain Name
To get a domain name, you can usually go through your ISP. However, not all
ISPs offer that service. The InterNIC Accredited Registrar Directory lists domain
No
name registration service providers from around the world that provide domain
c

name registration services. In that case, you must apply for a name through For more information, visit
InterNIC under the top-level domains (listed in the table above). the InterNIC Frequently
Asked Questions Web site at
For a list of international places at which you can register a DNS domain, visit http://rs.internic.net/faq.html.
en

http://www.yahoo.com/Computers_and_Internet/Internet/ Domain_Registration/
Do

Network_Information_Centers/
When applying for a domain name outside of the United States, complete
InterNIC’s template, but email it to iana@iana.org.
se
r

TASK 10B-1:
fe
ea

Discussing DNS
1. What is DNS?
Pl
Re

2. What are some examples of top-level domains?

Lesson 10: TCP/IP Fundamentals 237

 3. Create a fully qualified domain name for the R&D department of the
GNS corporation.

4. How can you get a domain name?

ial
Host Addresses

er
The host portion of the IP address is used to uniquely identify each computer on
a network segment, just as your house number uniquely identifies your house on
your street. The host address must be unique for each network address. For
example, you could use 130.4.67.8 and 130.4.45.34 for two hosts on your net-
work; they have the same network address, but unique host addresses. All TCP/
IP-based systems must have unique host addresses, including all computers,

at
network cards within routers, and printers with network cards.
Figure 10-10 shows a network connected by two routers. Notice that each com-

y
puter has its own unique host address.

op
eM
tC
No
c
en
Do
se
r
fe
ea

Figure 10-10: A network with multiple segments and with unique IP addresses assigned.
Pl
Re

238
Private Network Addressing
If you have a private network (no registered addresses), you can use the follow-
ing address blocks reserved by the InterNIC for private networks:
• Class A: 10.0.0.0 through 10.255.255.255
Remember that although the
• Class B: 172.16.0.0 through 172.31.255.255 host address portion of the

ial
IP address must be unique,
• Class C: 192.168.0.0 through 192.168.255.255 the network address portion
must be the same for all
hosts on the same segment
of the network.
The Default Gateway Address
The host address assigned to an IP router’s network card is particularly important

er
because this address is configured as the default gateway for all computers on
Direct Internet connections
that network segment. The default gateway address is the network segment’s
require registered addresses.
access point to all other remote networks.

at
TASK 10B-2: TCP/IP uses the terms
routers and gateways
interchangeably.

y
Discussing methods for assigning Host Addresses

op
1.
eM
What is the purpose of the host portion of the IP address?
default gateway address:
The network segment’s
access point to all other

tC
2. If your IP address is 175.135.200.1, what must the IP address of the net- remote networks. The default
work card in the router be? gateway address is the IP
address of the router that the
system will use to connect to
the other remote networks.
No
c

Rules for Host Addressing

en

You should use the following guidelines when assigning the host portion of an IP
Do

address:
• Each host address must be unique to the local network.
• The host address bits cannot all be 0. If all host bits are 0, then the resulting
IP address is the network address. For example, if the network address is
se
r

100 (Class A), and the host bits are all 0, then the resulting IP address is
100.0.0.0. This IP address is invalid because it is reserved to determine the
fe

network address for routing purposes.

•
ea

The host address bits cannot be all 1s (a decimal value of 255 for each
byte). If all host address bytes are set to 255, the transmission is thought to
be a broadcast.
Pl
Re

Lesson 10: TCP/IP Fundamentals 239

 The following table outlines the range of valid network addresses.

Network class Starting host address Ending host address

The first byte(s) of host
addresses in Class A and B A net.0.0.1 net.255.255.254
networks can be any value B net.net.0.1 net.net.255.254
from 0 to 255. For example, C net.net.net.1 net.net.net.254

ial
a valid Class B host address
can be 255.1.

TASK 10B-3:

er
Determining Valid Host Addresses
1. Is 221.100.155.0 a valid IP address? Explain why.

at
2. Is 187.131.255.5 a valid IP address? Explain why.

y
op
eM
3. Is 121.0.0.1 a valid IP address? Explain why.

tC
No
Guidelines for Assigning Host Addresses
c

While there are no rules for assigning specific addresses to different types of
computers, you can make assigning host addresses much easier if you establish
en

some guidelines. For example, you may find it easier to remember the function of
Do

a specific machine if you group the host addresses by type. If you use one of the
bytes from the host address in a Class A or Class B network, you could assign a
specific number to file servers or UNIX hosts. Consider the examples presented
in the following table.

Network
se
r

class IP address Example

A net.host.host.host Use net.101.host.host to identify file servers. Use net.
fe

201.host.host to identify UNIX hosts or printers with

ea

network cards.
B net.net.host.host Use net.net.101.host to identify file servers. Use net.net.
201.host to identify UNIX hosts or printers with network
Pl

cards.
Re

You could even assign specific ranges of numbers for the second or third byte of
the address, based on the type of operating system a computer runs. For example,
you could assign 51 to Windows 95 machines, 52 to Windows NT workstations,
101 to Windows NT servers, 102 to NetWare servers, and 201 to UNIX hosts.

240
You might also choose to assign values for the second or third byte of an IP
address based on a machine’s geographic location. For example, an address could
be used to identify hosts on different floors in your building.
Use the rest of the host address bytes to identify workstations and routers. For
example, you could use host addresses from 1 to 225 for workstations and from
225 to 254 for routers.

ial
By using guidelines to assign your IP addresses, you can get more information
from an IP address than just the network segment a computer is located on. For
example, if your network address is 154.131 and the host address is 52.1, then
the above guidelines would tell you that this computer is located on segment 154.
131 and runs Windows NT Workstation. In contrast, an address of 154.131.101.
254 would tell you that this computer is located on segment 154.131, runs

er
Windows NT Server, and functions as an IP router.

TASK 10B-4:

at
Identifying guidelines for assigning the Host portion
of IP addresses

y
op
1.
eM
What types of systems on your network require host addresses?

tC
2. What are some guidelines you have implemented or are thinking of
implementing for assigning host addresses on your own network?
No
c
en
Do

Host Tables
In IP networks, host names are translated into IP addresses on the Internet by
either Domain Name System (DNS) or a host table. DNS is a service used to
resolve domain names to addresses on the Internet. A host table is a network
directory containing a list, in ASCII file format, of commonly accessed names
se
r

and addresses. Each node has its own host table from which network host and
service addresses are resolved to names. host table:
fe

An ASCII text file that

A host table displays information in a single line, as shown in the following host
ea

contains the host names and

table. the associated IP addresses
for any host that is
commonly accessed on a
122.53.37.7 Workstation1 Lab1
TCP/IP network.
Pl
Re

43.09.53.56 Workstation2 Lab2

88.34.44.2 Workstation3 Lab3
12.64.66.9 Workstation4 Lab4
127.0.0.1 localhost
#New information will be added periodically

Lesson 10: TCP/IP Fundamentals 241

 Each line contains an IP address, host name, and any aliases, separated by the
desired number of [Tab] characters. You can insert comments into the table by
putting the pound sign (#) in front of your remark. Lines preceded by pound
signs are ignored when the table is searched.
When you attempt to establish a connection over the Internet, your system reads
through its host table to find the appropriate IP address for the connection. For

ial
example, if you wanted to connect to novell.com, your system would check its
host table to find an IP address for novell.com.
Host tables are useful in small networks where you can easily maintain copies of
On a NetWare server, the the tables on each system which needs the host table to resolve names to
host file is located in SYS:\ addresses. Can you imagine how huge this table would be if you managed host
ETC. On a UNIX server, it is tables for Internet access? It would be a full time job modifying and copying the

er
located in /etc/hosts. On an tables, and would take so much room on the systems, that it really wouldn’t
NT server, the host file is work. That’s why DNS is used on most networks today.
located in C:\Windows.

The LMHOSTS File

at
The LMHOSTS file is used to resolve network addresses to NetBIOS node names
(the computer name) on Microsoft-based systems. The file is found in the

y
C:\Windows directory. This is a static file that must be manually updated. It maps
NetBIOS names to the corresponding IP address. This name resolution method

op
eM
has been replaced by WINS.

tC
DNS Zones
A DNS zone is a section of the DNS database that contains domains. Represent-
ing a logical portion of the Internet, a zone expands downward in a DNS
hierarchical tree until it reaches either another domain or an end node, as dis-
No
played in Figure 10-12. Although a zone represents a logical Internet section, it
c

does not have to correspond to physical Internet segments. For example, a

domain can include multiple networks, or one network can span several domains.
en
Do
se
r
fe
ea
Pl
Re

242
 ial
er
at
y
Figure 10-11: A DNS zone.

op
eM
Resolvers
Resolver is another name for a DNS client that submits requests for IP addresses.

tC
The resolver sends the request to a local domain server, which checks its
database. If the server cannot find an address within its domain, it forwards the
request outside the domain to a root name server. Resolution of a name to an IP
address is seamless to a resolver; it sends only one request to its local server and
No
cannot distinguish whether the resolution comes from that server or one outside
c

the domain.
The process works much like making a directory assistance telephone call; you
dial directory assistance (for example, dial 411) and an operator helps you find
en

the phone number you need. You are not aware of what he or she does to get
Do

your information. resolver:

A DNS client that submits
requests to the local domain
server for IP addresses.
se
r
fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 243

 ial
er
at
y
op
eM
Figure 10-12: A resolver.
For example, if you want to connect to zdu.com, your computer sends a request
for zdu.com’s IP address to your DNS server, as shown in Figure 10-12. Your
DNS server tries to resolve the name, but if it can’t, it asks the root (.) name

tC
server, and then all other servers, from the requested domain. The server that can
resolve the name sends the IP address back to your DNS server, which sends it
back to your computer. The DNS server stores IP address requests in cache, so if
that same address is requested again, the server takes the information from cache
No
instead of searching for it.
c

TASK 10B-5:
en
Do

Discussing host tables and DNS zones

1. Compare using host tables and DNS.
se
r
fe
ea
Pl
Re

244
2. What is a DNS zone?

3. What is another name for a DNS client that submits requests for IP
addresses?

ial
Name Resolution

er
Let’s summarize the name resolution methods we’ve examined in this lesson. We
looked at static name resolution methods, such as using the HOSTS or
LMHOSTS files. We also looked at the dynamic name resolution methods, DNS
and WINS.
Let’s look at a video presentation that describes name resolution.

at
TASK 10B-6:

y
op
Name resolution
eM
Objective: To view a video presentation on the name resolution process.

tC
Setup: You’ll need your Certification CD for this task.

1. Load the Digital Video component of the Network+ Certification

CD-ROM.
No
2. Click on the Network Protocols link, and then click on the Overview
c

link to start the video presentation.

3. After viewing the presentation, list at least two ideas presented in the
en

video.
Do

4. Did this video presentation help you understand DNS and name resolu-
se
r

tion? Why or why not?

fe

5. Exit the Network+ Certification CD-ROM.

ea

Addressing Protocols
Pl
Re

The protocols listed in the following table are associated with the Address layer
of the DOD model.

Lesson 10: TCP/IP Fundamentals 245

 Name RFC Description
Address Resolution 826 Translates a host’s software address to a hardware (or MAC)
Protocol (ARP) address (the node address that is set on the network interface
card).
Reverse Address 903 Adapted from the ARP protocol and provides reverse
Resolution Protocol functionality. It determines a software address from a hardware

ial
(RARP) (or MAC) address. A diskless workstation uses this protocol
during bootup to determine its IP address.
BOOTP 1542 Also used by diskless workstations. It enables these types of
workstations to discover their IP addresses, the address of a
server host, and the name of the file that should be loaded into
memory and run at bootup.

er
ARP and RARP

at
A frame’s MAC header includes only a hardware address, which causes no prob-
lem if the packet is traveling to a destination on the network from which it was
sent. However, if it is traveling to a different network, the packet must be identi-

y
fied with an IP address.

op
eM
To associate a packet with an IP address by using a hardware address, TCP/IP
uses the Address Resolution Protocol (ARP). The most universally used address
resolution method, ARP maps 4-byte software-based IP addresses to 6-byte
hardware-based addresses in the Data Link layer. Conversely, Reverse Address

tC
Resolution Protocol (RARP) extends ARP’s capabilities by mapping a hardware
address to a device’s IP address. RARP is extremely useful in finding network
addresses of diskless terminals.
To do its job, each system’s ARP maintains a table of IP addresses mapped to
No
specific hardware addresses. When a packet travels between networks, ARP
c

checks its table for the appropriate IP address. If the address is not in the table,
ARP issues a broadcast request to all network hosts asking for the address, as
displayed in the following figure. All active hosts and routers supporting ARP
en

hear the request and any device aware of the address adds it to the ARP table.
Do
se
r
fe
ea
Pl
Re

Figure 10-13: ARP and RARP.

246
You do not need an IP router for MAC addressing. However, an IP router can
respond to an ARP request because ARP is automatically enabled on a product
that supports it. An IP router can also respond to ARP requests for a host on
another subnetwork, a service known as Proxy ARP.

ial
BOOTP
If a host does not know its own IP address, it contacts the BOOTP server, which
keeps a list of each host’s logical and physical addresses. The BOOTP server
looks up the host in its table and returns the appropriate information to the
requesting host. Similar to ARP, an IP router is used when the BOOTP server
lives on a different network than the requesting host. In this situation, the host

er
tells the IP router when it must forward BOOTP requests to available BOOTP
servers. For the IP router to do so, it must know BOOTP server addresses
throughout the internetwork.
BOOTP and RARP serve the same purpose. BOOTP is a newer, more frequently-

at
used version of RARP.

y
Dynamic Host Configuration Protocol (DHCP)

op
eM
One of the supporting protocols incorporated into TCP/IP is the Dynamic Host
Configuration Protocol (DHCP). DHCP is used to centrally administer the assign-
ment of IP addresses, as well as other configuration information such as subnet
masks and the address of the default gateway. When you use DHCP on a TCP/IP

tC
network, IP addresses are assigned to clients dynamically instead of manually.
A DHCP server assigns dynamic IP addressing information to hosts for a period
of time referred to as a lease. Using a lease in a dynamically configured environ-
ment has several advantages. New configuration parameters, such as a new
No
default gateway address, can be assigned to hosts when the lease is renewed.
c

Also, IP addresses that are no longer in use are returned to the DHCP server for
assignment to other hosts.
en

You can also specify either automatic or manual allocation of IP addresses to

Do

hosts. With automatic allocation, the DHCP server assigns a host an IP address.
With manual allocation, you specify an address for a particular host and the
DHCP server simply delivers it.
As mentioned, DHCP takes care of assigning IP addresses to hosts so that you do
not have to. The DHCP process includes the following steps.
se
r

1. Upon booting, a host requests configuration information via a broadcast

packet.
fe

2. The DHCP server hears the broadcast and checks the packet to find out
ea

which network it came from.

3. The DHCP server returns the appropriate information to the host.
Pl
Re

If the DHCP server does not

have configuration
information for the requested
network segment, it simply
drops the request.

Lesson 10: TCP/IP Fundamentals 247

 Comparing Manual and Dynamic TCP/IP Configuration
You can manually configure IP addresses for your systems, or you can have
DHCP dynamically assign IP addresses for the systems on your network. Let’s
complete a couple of simulations; one will have you manually assign a TCP/IP
address and in the second, you’ll configure the workstation to dynamically
receive an IP address.

ial
TASK 10B-7:
Installing and configuring TCP/IP

er
Objective: To complete two simulations in which you’ll install and con-
figure TCP/IP manually and dynamically.
Setup: This task requires use of the Network+ Certification CD-ROM.

at
1. Load the Activities component of the Network+ Certification CD-ROM.

2. Click on the Network Protocols link, and then click on the TCP/IP Fun-

y
damentals link.

op
3.
eM
Point to the Install And Configure TCP/IP Manually link. Read through
the scenario, and then click on the link.

4. Complete the steps in the Activity as prompted in the lower-right corner

tC
of the screen, being sure to complete them in the simulation, and not on
your actual workstation! Click the Exit button to return to the Activities
list.
No
5. Point to the Install And Configure TCP/IP To Use DHCP link. Read
c

through the scenario, and then click on the link.

6. Complete the steps in the Activity as prompted.

en
Do

7. When you finish the Activity, exit the Network+ Certification CD-ROM.

Relay Agents
se

The purpose of the DHCP relay agent is to enable hosts to obtain IP addresses
r

from DHCP servers located on other subnets. Routers cannot forward broadcast
packets, so when a DHCP host on a subnet containing a DHCP relay agent but
fe

relay agent: no DHCP server requests an IP address, the request is forwarded to the DHCP
ea

Enables hosts to obtain IP relay agent. The DHCP relay agent, in turn, is configured to forward the request
addresses from DHCP directly to the DHCP server on the other subnet. The DHCP server then returns
servers located on other
subnets.
the IP address directly to the DHCP host.
Pl
Re

The DHCP relay agent must be configured with the IP address of the DHCP
server, so the relay agent knows where to forward requests from DHCP hosts
needing IP addresses.

248
Host-to-Host Layer Protocols
At the Host-to-Host layer, the protocols used are Transmission Control Protocol
(TCP) and User Datagram Protocol (UDP).

Remember that the Host-to-

Host layer on the DOD

ial
model is equivalent to the
OSI Transport layer.

er
at
y
op
eM
Figure 10-14: Host-to-Host layer protocols.

tC
The following table describes the Host-to-Host layer protocols.

Name RFC Description

Transmission Control 793 Provides for data exchange between dissimilar hosts by
No
Protocol (TCP) using acknowledged, full-duplex, connection-oriented,
c
flow-controlled, and sequenced communication.
User Datagram Protocol 768 Like TCP, provides delivery service. However, UDP does
(UDP) not guarantee or acknowledge the delivery; it only
en

accepts and transports Upper Layer Protocol (ULP)

Do

datagrams.

TCP is similar to a telephone

TCP call or registered mail. UDP
se
r

is similar to a television
TCP provides reliable connections that are provided by virtual circuits. It also broadcast or regular mail.
supports multiple simultaneous ULP conversations. TCP can be compared to reg-
fe

istered mail. When you send a letter by registered mail (with return receipt), you
ea

always know whether your letter reached the addressee; with TCP, every message
sent is acknowledged at the destination. Telnet, FTP, HTTP, and SMTP are
examples of applications and protocols that use the TCP protocol.
Pl
Re

Lesson 10: TCP/IP Fundamentals 249

 ial
er
at
y
op
eM
tC
Figure 10-15: The TCP header.

UDP
No
The advantage of UDP is that it lets applications send datagrams without the
c

overhead involved in acknowledging packets and maintaining a virtual circuit.

UDP is therefore used to broadcast messages across an internetwork, because
acknowledgment is unnecessary and overhead is undesirable.
en
Do

This protocol can be compared to regular mail. As with mailing a letter, you have
no way of really knowing if, or when, the letter arrives at its destination.
se
r
fe
ea
Pl
Re

Figure 10-16: The UDP header.

250
Because UDP does not acknowledge packets, applications that use it accept full
responsibility for maintaining reliable communication and preventing message
loss, duplication, delay, out-of-order delivery, and so on. Like TCP, a UDP header
contains source port, destination port number, and checksum fields. However, a
UDP header is only 16 bits and contains a UDP length field that specifies the
number of bytes in the whole UDP datagram.

ial
NFS and TFTP are examples of applications that use the UDP protocol. Other
protocols that rely on broadcasts also use it.

Process/Application Layer Protocols

er
TCP/IP applications provide users with services such as terminal emulation, file
transfer capability, and electronic mail.
TCP/IP applications use the client-server communication model. That is, each Most TCP/IP applications are
application contains two parts: a client program that sends a request to a server, more similar to NOS utilities
and a server program that fulfills the request sent by the client. The client pro- than they are to desktop

at
gram is executed on a user’s workstation to gain access to the server. The server applications such as word
program is loaded automatically and runs as a background process. processors or spreadsheets.

y
op
eM A program that is loaded and
running as a background
process on the server is
referred to as a daemon.

tC
No
c
en
Do
se

Figure 10-17: Process/Application layer protocols.

r

The following table describes some common Process/Application layer protocols.

fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 251

 Name RFC Description
File Transfer Protocol 959 Enables the transfer of files between a user’s workstation
(FTP) and a remote host. Using FTP, virtual circuits establish
reliable connections between hosts. A user can access
the directory structure on a remote system, change
directories, search and rename files and directories, and

ial
download and upload files. Users must have an account
and use a password for file transfers unless they are
configured for anonymous access. In order to enable a
NetWare client to access the directory structure on the
TCP/IP host server, TCP/IP support must be installed on
a client.

er
Trivial File Transfer 1350 Enables a user to transfer files between a workstation and
Protocol (TFTP) a host without specifying an account or password.
Because the protocol does not use accounts and
passwords, access to most parts of a host server’s
directory structure will be denied. A TFTP user will be
able to download only public files to which everyone has

at
access. TFTP runs on UDP and therefore does not
require a reliable packet delivery method.
Post Office Protocol 3 1725 Receives and stores email for users at their Internet mail

y
(POP3) server. POP3 allows a workstation to retrieve mail that the
server is holding for it.

op
eM
Simple Mail Transfer 821 Provides electronic mail (email) services, enabling users
Protocol (SMTP) to send and receive electronic messages. It does not
provide the mail interface. Rather, it specifies how

tC
messages are transferred. This type of application is the
most common on TCP/IP networks. Gateways, such as
an SMTP gateway, can be used to connect dissimilar
messaging systems.
HyperText Transfer Protocol 1945 Provides a client/server environment that enables
No
(HTTP) request/response communication with the WWW and
c

intranets. A client sends a request to a server*. It also

enables communication between gateways and agents
using different TCP/IP protocols, such as FTP and SMTP.
en

Telnet 854, 855 A terminal emulation protocol that enables a user at one
Do

site to simulate a session on a remote host. It does this

by translating keystrokes from the user’s terminal to
instructions recognized by the remote host, and then
carrying the output back to the user’s terminal and
displaying it in a format native to the remote host. This
se

service is called transparent, because it gives users the

r

impression that their terminals are directly attached to the

remote host.
fe

Network File System (NFS) 1813, 1094 Provides transparent remote access to shared files across
ea

networks. The NFS protocol is designed to be portable

across different machines, operating systems, network
architectures, and transport protocols.
Pl

Line Printer Daemon (LPD) 1179 Provides printing services. A host (LPD client) requests
Re

print services by issuing one of the commands

associated with line printer spooling, such as lpr (assign
to queue); a print server (LPD server) responds to the
request by performing the command.

252
Name RFC Description
Simple Network 1157 Provides remote management capabilities. On a network
Management Protocol running SNMP, an administrator can manage an entire
(SNMP) network from one station.

ial
HTTP requests consist of the following information:
• Message type. Simple-request and simple-response messages are limited to
the single request method called GET. However, full-request and full-
response messages can use many different types of methods, such as GET,
PUT, and POST.
• Method. The way in which the information will be retrieved.

er
• Identifier. A Uniform Resource Identifier (URI) indicating which resource
should act on the request.
• URL. A Uniform Resource Locator (URL) that identifies a network resource.
• Version. Which HTTP version is being used. HTTP/0.9 is used as a default.

at
SNMP

y
op
SNMP was designed to enable the analysis and troubleshooting of network
eM
hardware. For example, SNMP enables you to monitor workstations, servers,
minicomputers, and mainframes, as well as connectivity devices such as bridges,
routers, gateways, and wiring concentrators.

tC
The two required components for managing a network using SNMP are SNMP
Managers (or management systems) and SNMP Agents. SNMP Managers request
information from SNMP Agents concerning their current configuration, network
activity, and statistics on performance. Information requested of agents is located SNMP:
No
in a host’s Management Information Base (MIB). The MIB is a database of infor- Simple Network Management
c

mation about defined objects, such as physical and network addresses, statistics, Protocol. The Internet
counters, and route and ARP tables. An example is the LAN Manager MIB, standard protocol developed
which has objects for collecting information on user and logon statistics. to manage nodes on an IP
network. It can be used to
en

SNMP Managers retrieve information from SNMP Agents by using specific com- monitor and network devices
Do

mands or operations. The get operation enables the SNMP Manager to request a and their functions.
specific value; for example, an SNMP Manager can issue a get command to a
DHCP server configured as an SNMP Agent to obtain its current number of
active leases. The get-next operation is used to request the next value from a
defined set of values. The get-response operation is used to fulfill a request by the MIB:
se

Management Information
r

manager. The set operation enables the SNMP Manager to change the value of a Base. A database of
parameter on the SNMP Agent. information about defined
fe

objects, such as physical

ea

and network addresses,

statistics, counters, and route
and ARP tables.
Pl
Re

Not all vendors support all

MIB objects in their SNMP
implementations.

Lesson 10: TCP/IP Fundamentals 253

 An SNMP Agent performs the commands requesting information issued by the
SNMP Manager. SNMP Agents can also send alarms, or traps, to the SNMP
Manager if network errors or configurable significant events occur. An example of
Trap messages are a significant event would be network traffic exceeding a certain number of pack-
essentially warnings; they ets per second. You can configure some SNMP Agents, such as a router, to send a
inform a manager when
trap to the SNMP Manager if the traffic exceeds a threshold value. SNMP Agent
certain thresholds are met or
software is available for most operating systems and network devices. The fol-

ial
if there is a problem.
lowing table describes situations in which an SNMP agent would send a trap.

Trap situation Sent because

Link up or down IP has failed or come back up. (Trap sent by a server.)
EGP neighbor loss Communication between an SNMP agent and its Exterior Gateway

er
Protocol neighbor fails.
Warm/cold boot The current configuration might need its tables reinitialized.
Failure to authenticate An SNMP request includes an unrecognized community name. (Trap
sent by a server.)

at
Depending on a vendor’s IP implementation, an IP application can issue trap
messages when events other than those listed occur. By default, a trap is sent to
your local host address (127.0.0.1). However, you can specify a different

y
destination.

op
eM
For security reasons, you can configure SNMP Managers and Agents into com-
munity names. A community name is a case-sensitive text string with a limit of
32 characters that acts like a password. The SNMP Manager sends it with every
agent request. An SNMP Agent can communicate only with SNMP Managers

tC
within its community; however, an SNMP Agent can be a member of more than
one community at a time. Because SNMP Managers can retrieve sensitive infor-
mation about your network, restricting SNMP communications to within
communities helps to prevent others from placing their own SNMP Managers on
No
your network and retrieving information about it. The installation of SNMP
c

Agents defaults to a community name of Public. The following list displays a

fictitious set of community names.
• ″ENGINEERING″
en

• ″PRODANDDIST″
Do

• ″RESANDDEV″

Ports and Sockets

se
r

Service providers (file servers, print servers, application servers, and so forth) and
service requesters (clients) find out about each other through broadcasts or other
fe

means. This stage of communication lets the client and server determine each
ea

other’s node addresses. Once a service is needed, a communication channel is

required. Such a communication channel is made possible by identifying informa-
tion that each party to the communication knows about the other. Such
Pl

identifying information consists of the node addresses of each station and identifi-
Re

ers that describe what programs on the client and server are communicating.
Successful communication employs conversations between service requesters and
clients through addressing.

254
A common addressing method uses connection IDs, also called a connection iden-
tifier, socket, or port. These usually apply to a program on the server. For
example, a Web server is typically configured to respond to communications
using port 80. However, all clients communicate simultaneously with that same
port. You could think of a port as being the address of an application on a
computer.

ial
When a computer receives a packet destined for a particular port, that packet is
given to the application running the service associated with that port. Some com-
monly used ports (also known as Well Known Ports) include:

Application Port
FTP 21

er
Telnet 23
SMTP 25
DNS 53
TFTP 69

at
finger 79
HTTP (Web) 80
POP3 110

y
NNTP 119

op
SNMP
eM 161

tC
RFC 1700, Assigned
Analysis Tools Numbers, contains a
comprehensive list of Well
You might want to implement protocol analyzers, such as NetWare LANalyzer (a Known Port addresses.
ManageWise agent), Sniffer, and NCC LANalyzer, on your network to monitor
No
performance. At the OSI Network layer, protocol analyzers figure out distances
c

between servers and clients to help you discover inefficiencies in your cable
configurations. At the OSI upper layers (Transport, Session, Presentation, and
Application), analyzers can examine an application during an application down-
load between a client and server.
en
Do

TASK 10B-8:
Reviewing the Components of the TCP/IP Protocol
se
r

Suite
fe

1. What is a daemon?
ea

2. Which protocol provides terminal-emulation functionality?

Pl
Re

3. Which protocol provides email functionality?

Lesson 10: TCP/IP Fundamentals 255

 4. Which protocols enable you to transfer files from host to host?

5. What is the main difference between TCP and UDP?

ial
6. Name an application that uses TCP.

7. Name an application that uses UDP.

er
8. Which protocol deals with error and control messages?

at
9. Which protocol is responsible for packet fragmentation, routing and
assembly?

y
10. Which two protocols are used by diskless workstations?

op
eM
11. Which protocol translates software addresses to MAC addresses?

tC
12. What types of specifications are defined at the Network Access layer?
No
c

13. Identify the following data unit names with the appropriate DOD layers:
frame, packet, data, and message.
en
Do

APPLY YOUR KNOWLEDGE 10-1

se

Suggested time:
r

20 minutes In this activity, you’ll work through some sample test questions that deal with the
fe

information presented in this lesson.

ea

1. Load the Skill Assessment component of the Network+ Certification

CD-ROM. (Remember to accept the disclaimer.)
2. Start the Network+ test.
Pl
Re

3. Select all items for the TCP/IP Fundamentals topic, uncheck the Test Time
check box, and then click OK.
4. Work through the sample questions and score the test.
5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

256
6. Exit the test.

Summary

ial
In this lesson, you have examined the basics of the TCP/IP protocol suite.
You looked at where it came from, why it is used, and examined several of
the most commonly used protocols contained within the TCP/IP protocol
suite.

er
LESSON 10 REVIEW
10A Describe how TCP/IP communicates information.

at
y
op
eM
Where did TCP/IP originate, and what was its purpose?

tC
10B List several of the well-known ports and the application related to that
port.
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 10: TCP/IP Fundamentals 257

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
IP Addressing, Subnetting, LESSON
and Supernetting
11

ial
Data Files:
Network+ Certification
Overview CD-ROM

er
In this lesson, you’ll examine how IP addresses are constructed and examine Lesson Time:
how subnetting and supernetting are used with IP addresses. 2 hours

at
Objectives
To understand the use of IP addresses, you will:

y
11A Identify how IP uses an addressing scheme to connect computer

op
networks.
eM
In this topic, you’ll be identifying the components of an IP address,
examining the four IP address classes, and identifying enhancements

tC
included in IPv6.

11B Examine interior gateway protocols.

This topic discusses distance vector routing, split horizon, poison reverse,
No
link-state routing, and OSPF.
c

11C Begin working with subnetting.

Network segments on TCP/IP networks are identified by subnetting each
en

segment from the network’s single IP address. This topic shows you how
Do

to implement subnetting on your network using default subnet masks.

11D Examine the use of supernetting.

In an attempt to help prevent the depletion of network addresses avail-
able, supernetting or classless inter-domain routing has been developed.
se
r
fe
ea
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 259

 TOPIC 11A
IP Addressing
To enable communications on a network, each protocol suite must employ an

ial
addressing scheme for its network and devices. For example, the IP uses a 32-bit
combination of network and host IDs as addresses. A host is a workstation or a
server node (or station). By using a host’s IP address, a router can find the host
on the Internet.
A host’s IP address is made up of a network address and a host address. The net-
work address uniquely identifies each segment of a network. The host address

er
uniquely identifies each machine on a segment. If your network is private (not
connected to the Internet), you can design any addressing scheme you want. But
if you want to connect to the Internet, you must apply to the Internet Network
Information Center (InterNIC) for an address. Internet addresses are centrally
maintained to guarantee that no addresses are duplicated.

at
y
op
eM
tC
No
c
en
Do

Figure 11-1: A breakdown of the IP addressing scheme.

IP addresses have 4 bytes, each of which is referred to as an octet . Since each
byte in the address has 8 bits, an IP address is 32 bits long. IP addresses are usu-
se

IPv6 enables the use of 16-

r

byte IP addresses. ally displayed in decimal format where the value of each byte is converted from
binary to decimal. This makes them easier to remember. For example, an IP
fe

address of 198.131.25.4 is much easier to remember than its binary equivalent of:
ea

11000110.10000011.00011001.00000100
Pl
Re

260
The binary numbering system is based on 2, whereas the decimal numbering sys-
tem is based on 10. The place columns in the binary system are powers of 2;
thus, the right-most digit in a binary number is equivalent to 20, or 1; the next
digit is 21, or 2. Working from right to left, just as you would in the decimal sys-
tem, the placeholders are 1, 2, 4, 8, 16, 32, 64, and 128. Similarly, in the decimal
system, the placeholders from right to left are 1, 10, 100, 1000, and so on.

ial
By knowing that the placeholders in binary convert to powers of 2, you can con-
vert bytes to their decimal values easily. The following table shows how to
convert a binary number into a decimal number.

Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1

Decimal 27 = 26 = 64 25 = 32 24 = 16 23 = 8 22 = 4 21 = 2 20 = 1

er
value of 128
placeholder
Binary value 1 1 0 0 1 0 0 1
of byte
Binary to 128 64 0 0 8 0 0 1

at
decimal
conversion
Decimal 128+64+8+1

y
value of byte = 201

op
eM
Now, it’s your turn to convert some binary values to decimal.

TASK 11A-1:
Converting IP Addresses from Binary to Decimal Values tC
No
Objective: Convert the following IP addresses from binary numbers to
c

their equivalent decimal values.

1. 01100100
en
Do

2. 11100000
se
r

3. 11111111.11111111.11110000.00000000
fe
ea

4. 01100100.01100100.00000010.00000001
Pl
Re

5. 011111111.00000000.00000000.00000001

Lesson 11: IP Addressing, Subnetting, and Supernetting 261

 Converting IP Addresses from Decimal to Binary Values
Now that you’ve seen how to convert binary values to decimal, let’s look at how
you convert numbers from decimal to binary values.
Make the chart as you would for binary to decimal conversion:

ial
Bit 8 Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1
Decimal 27 = 26 = 64 25 = 32 24 = 16 23 = 8 22 = 4 21 = 2 20 = 1
value of 128
placeholder
Binary value 1 1 0 0 1 0 0 1
of byte

er
Binary to 128 64 0 0 8 0 0 1
decimal
conversion
Decimal 128+64+8+1
value of byte = 201

at
Starting with the leftmost column that isn’t greater than the decimal number

y
you’re trying to convert to binary, enter a one, then move to the next column. If
you add the value of that column to the first column, is the result greater than the

op
eM
number you’re converting? If so, enter a zero and move to the next column and
try it there. In the example above, to convert 201 to binary, you can use columns
128 and 64, but then you need to skip down to the 8 column, and then skip down
to the 1 column. Place zeros in any column you don’t use in the calculation to

tC
create your eight-bit binary number.

TASK 11A-2:
No
c

Converting IP Addresses from Decimal to Binary Values

Objective: Convert the following IP addresses from decimal values to
en

their equivalent binary numbers.

Do

1. 255
se

2. 8
r
fe
ea

3. 198.131.205.5
Pl

4. 224.254.100.8
Re

262
The Internet was designed to support three categories of networks:
• Those consisting of a small number of networks with many hosts;
• Those consisting of a medium number of networks with not quite as many
hosts; and

ial
• Those consisting of a large number of networks with a relatively small num-
ber of hosts.
To sustain these different types of networks, the Internet was designed to support
classes of IP addresses. These classes are implemented by using varying portions
of the four bytes of the IP address to identify networks and hosts. For example, a

er
Class A IP address uses only the first byte of the IP address to identify the net-
work address and the remaining three bytes to identify hosts. In contrast, a Class
B IP address uses the first two bytes of the IP address to identify the network and
the remaining two bytes to identify hosts. Thus, a Class A IP address supports
only a small number of network addresses and a large number of hosts, while a

at
Class B IP address supports a larger number of network addresses but not as
many hosts as a Class A. The Internet was designed to support five classes of IP
addresses, A through E. Most TCP/IP implementations support only classes A, B,

y
and C for host addressing.
The number of bits reserved for the network and host IDs is different for classes

op
eM
A, B, and C. The first bit(s) of the first byte of the IP address are specific for
each address class. Since the first few bits of an address are specified by the
InterNIC, the decimal value of the first byte can be used to identify the class of
IP address. The following table explains the rules for calculating the network

tC
addresses for each network class.

Range of
Value of high Range of values values for
No
Address order bit(s) in first for first byte in first byte in
c

class byte binary decimal Example

A First bit must be 0 00000001 - 1 to 127 (127 is 1.net.host.host to
01111111 Note: A reserved for 127.net.host.host
en

value of 0 for the testing purposes)

Do

network address is
not permitted.
B First 2 bits must be 10000000 - 128 to 191 128.net.host.host
10 10111111 to 191.net.host.
host
se
r

C First 3 bits must be 11000000 - 192 to 223 192.net.net.host to

110 11011111 223.net.net.host
fe

D First 4 bits must be 11100000 - 224 to 239 This range of

ea

1110 11101111 addresses is

reserved for
multicasts and is
not supported for
Pl
Re

host addressing.

Lesson 11: IP Addressing, Subnetting, and Supernetting 263

 Range of
Value of high Range of values values for
Address order bit(s) in first for first byte in first byte in
class byte binary decimal Example
E First 5 bits must be 11110000 - 240 to 255 This range of
11110 11110111 addresses is

ial
reserved for
experimental
purposes and is
not supported for
host addressing.

er
Some of the IP addresses have special uses and cannot be assigned to networks
and hosts. For example, the IP address 127.0.0.1 is reserved for testing purposes.
It identifies you—your network and host. The command ping 127.0.0.1 is used to
determine if TCP/IP is functioning properly. You use it to ping yourself. The fol-
lowing chart describes IP addressing restrictions and the reasons behind them.

at
Restriction Reason Example
Microsoft’s multicasting

y
product, NetShow, will A network address of 0 When the network address is set to 0, 0.0.0.22 (This address
temporarily replace a host’s (binary 00000000) is not TCP/IP interprets the IP address as a identifies host 22 on the local

op
″local″ address, meaning that the data
assigned address with a
dynamic Class D address.
permitted.
eM packet does not need to be transmitted
network.)

through a router.
A host address of 0 (binary When the host address is set to 0, 122.0.0.0 (This address

tC
00000000) is not permitted. TCP/IP interprets the address as identifies the network segment
identifying the network. 122.)
The network address of 127 Messages addressed to a network 127.0.0.1 (This address is
is reserved. address of 127 are not transmitted out referred to as the loopback
No
onto the network; these messages are address.)
reflected back to the transmitting
c

machine. The address of 127 is used

to test the configuration of TCP/IP.
Neither the network address The 255 address is reserved for 255.255.255.255 (This
en

nor the host address can be broadcasts. broadcast will be sent to all
Do

255. hosts on all networks.)

187.205.255.255 (This
broadcast will be sent to all
hosts on network 187.205.)
A network address of all 1s
se

TCP/IP identifies all hosts with that 1.1.1.1 (This address refers to
r

is not permitted. address. every host.)

fe
ea

The formula for calculating the available number of network or host addresses for
each of the address classes is 2 raised to the power of bits available for varying.
For example, Class A network addresses have 7 bits available for varying minus
the7 unavailable addresses of 0 and 127; thus, the available number of networks is
Pl
Re

2 - 2 or 126. The following table defines the number of available network and
host addresses for the different network classes.

264
Network Number of Number of hosts per
class IP address networks network
A net.host.host.host 126: Formula = 27 - 2(0 16,777,214: Formula = 224 - 2
network = 7 bits and 127 are unavailable; (255 and 0 are unavailable.)
hosts = 24 bits 127 is reserved for
testing TCP/IP’s

ial
configuration.)
B net.net.host.host 16,384: Formula = 214 65,534: Formula = 216 - 2
network = 14 bits
hosts = 16 bits
C net.net.net.host 2,097,152: Formula = 254: Formula = 28 - 2
network = 21 bits 221

er
hosts = 8 bits

Note: Classes D and E are not discussed because they are reserved for certain
uses and cannot be assigned to hosts.

at
Obtaining an IP address from the InterNIC

y
If your network is strictly private and does not connect to the Internet, you can

op
eM
select any IP network address you want. However, if your network connects to
the Internet, you must apply to the InterNIC Registration Services for an IP
address. The InterNIC usually assigns large blocks of addresses to large Internet
Service Providers (ISPs). These ISPs distribute the addresses to smaller ISPs who

tC
then pass them on to users. InterNIC:
Internet Network Information
Center. An integrated network
information center developed
by Network Solutions, AT&T,
No
General Atomics, and the
c

U.S. Government. It provides

and coordinates services for
the Internet.
en
Do

You can visit InterNIC’s

online registration services at
http://internic.net. You can
also apply by email at
hostmaster@internic.net.
se
r

Alternatively, you can send

applications for IP addresses
fe

to Network Solutions,
InterNIC Registration
ea

Services, 505 Huntmar Park

Drive, Herndon, VA 22070
((703) 742-4777).
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 265

 The InterNIC will assign the network portion of an IP address to your company;
you usually will be able to select your own host addresses for the assigned net-
work address. Currently, only Class C addresses are available. All Class A
addresses have been assigned; only a few Class B addresses are available, and
they are reserved for large networks.
The American Registry for Internet Numbers (ARIN) is also used to assign IP

ial
ICANN (Internet Corporation address on behalf of the InterNIC. Both organizations manage addresses in North
for Assigned Names and and South America, South Africa, and the Caribbean. You can find more informa-
Numbers) is a non-profit tion about IP registration in your area by accessing the following URLs:
corporation responsible for
• http://www.arin.net/ for those countries listed previously;
allocating IP address space,
assigning protocol • http://www.ripe.net/ for European countries; and
parameters, DNS
•

er
management, and root server
http://www.apnic.net/ for Pacific Asian countries.
management functions. IANA
and other groups under U.S.
Government contracts
perform these functions. For Private Addressing
more information about IANA

at
(Internet Assigned Numbers
IANA has set aside three address ranges for use by private or non-Internet con-
Authority), visit http://www. nected networks. This is referred to as Private Address Space and is defined in
iana.org. RFC 1918.

y
The reserved address blocks are:

op
eM
10.0.0.0 to 10.255.255.255 (10/8 prefix)
ARIN: 172.16.0.0 to 172.31.255.255 (172.16/12 prefix)
American Registry for
Internet Numbers. Regional 192.168.0.0 to 192.168.255.255 (192.168/16 prefix)

tC
Internet registry for obtaining
IP addresses in the United If a private network does need to connect to the Internet (or other TCP/IP net-
States. works), you can use Network Address Translation (as described in RFC 1631).
The translation occurs on the router connecting the private network to the
Internet. The private addresses are translated into valid IP addresses and then the
No
packets are forwarded on to the outside world (beyond the private network).
c

RIPE:
Reseau IP Europeens.
Regional Internet registry for
obtaining IP addresses in
en

Europe.
Do

APNIC:
Asia Pacific Network
se

Information Center. Regional

r

Internet registry for obtaining

IP addresses in the Asia-
fe

Pacific region.
ea

Private Address Space is

Pl
Re

defined in RFC 1918.

266
IP Version 6
Because the number of available IP addresses is limited, as well as reasons
described below, a new version of IP, called IP Next Generation (IPng) or IP Ver- For more information on
sion 6 (IPv6) , is under development. IP Version 6 will enable the InterNIC to IPv6, visit http://www.iana.
assign a greater number of addresses; thus, the Internet will be able to support org/ipv6-announcement.txt,
more networks. The currently used IP header, known as Version 4, has not http://playground.sun.com/

ial
changed since the 1970s. The initial design did not anticipate the growth of the pub/ipng/HTML/INET-Ipng-
Internet and the eventual exhaustion of the IP Version 4 (IPv4) address space; Paper.HTML, or http://
playground.sun.com/pub/
therefore, a new version of IP is now needed. This new version incorporates the ipng/HTML/ipng-main.
ideas of many different proposed methods of creating a newer version of the IP HTML.
protocol. IP Version 6 is composed of an entirely new packet structure that is
incompatible with IP Version 4.

er
IP Version 6 contains the following features:
• Extended address space. IPv6 has 128-bit source and destination IP
addresses, making the addresses four times longer than in IPv4. This makes
plenty of addresses available in the foreseeable future. With IPv6, a valid IP

at
address may appear as a hexadecimal number, such as 4B2E: BD58: E241:
47C4: 3508: BC41: 331E: 1204.
• Simplified header format. The IPv6 headers are designed to keep the IP

y
header overhead as small as possible. Fields that are non-essential, as well as

op
option fields, have been moved to extension headers that are placed after the
IP header.
eM
• Support for time-dependent traffıc. A new field in the IPv6 header enables
the allocation of network resources along a path so that time-dependent ser-

tC
vices, such as voice and video, are guaranteed a requested bandwidth with a
fixed delay.
• Extensibility. IPV6 can be extended for unforeseen features through the addi-
tion of extension headers after the base header. Support for new hardware
No
and application technologies is built in.
c

TASK 11A-3:
en
Do

Determining IP Address Classes

1. If your network consisted of 200,000 individual computers worldwide,
what class of IP address would you need to directly connect each com-
se
r

puter to the Internet?

fe
ea

2. What is the network address of 127 used for?

Pl
Re

3. If you plan to connect your network to the Internet, what are the
restrictions for assigning your IP addresses?

Lesson 11: IP Addressing, Subnetting, and Supernetting 267

 4. How do you calculate the available number of host addresses for a Class
B address?

ial
5. Why is a host address of 255 not allowed?

6. Why is a host address of 0 not allowed?

er
TOPIC 11B

at
Interior Gateway Protocols

y
As mentioned previously, the Internet Protocol (IP) is responsible for transferring
data. It supplies the addresses and connectionless services used to forward data

op
eM
packets. However, it uses additional protocols for address resolution,
prioritization, and dynamic route discovery. IP also provides packet switching.
IPv6 also includes user authentication, 16-byte addressing, autoconfiguration, and

tC
backward compatibility with IPv4.

Autonomous System
No
IP is well suited for big, complicated internetworks because it enables the divi-
c

sion of an internetwork into logical groups called autonomous systems. An

autonomous system is a collection of networks (routers), administered by a single
authority that use the same Interior Gateway Protocol (IGP) to route packets. In
en

this case, router and gateway are interchangeable; routing protocols can be
Do

referred to as gateway protocols and vice versa. There are two common types of
IGPs: distance vector and link-stating.
Autonomous systems interconnect by using an exterior gateway protocol, such as
Exterior Gateway Protocol (EGP) and Border Gateway Protocol (BGP). EGP
se

enables communication between two end routers on different autonomous

r

systems. However, it lacks scalability and is best suited for small, simple
internetworks. BGP, an interdomain routing protocol, is an enhanced version of
fe

EGP and includes more scaling capabilities.

ea

Distance Vector Routing

Pl
Re

TCP/IP employs Routing Internet Protocol (RIP) as its distance vector routing
RIP is an implementation of protocol. Routers on interconnected LANs broadcast their distances from other
distance vector routing. LANs, as shown in Figure 11-2, which hosts use when deciding which way to
route their data.

268
Distance vector routing is similar to a train trip or an airplane flight. On a train
trip, you might need to switch trains at one or more railway stations. On an air-
plane flight, you might have layovers and plane changes. These correspond to
network routers. The shortest trip is often the one with the fewest changes; how-
ever, time is usually not the only consideration. You usually also consider
whether you need to pay a higher price for the shortest trip than you would for
another trip, whether you get frequent traveler discounts with another carrier, and

ial
the overall performance for the carrier. These factors correspond to the hop count.

er
at
y
op
eM
tC
Figure 11-2: Distance vector routing with RIP.
The following describes the process illustrated in Figure 11-2:
1. Router A broadcasts its requests, learns that it is 1 hop from LAN1, and
updates hosts on LAN2 with this information.
No
2. Router B broadcasts its requests and learns from hosts on LAN2 that Router
c

A is 1 hop from LAN1. It therefore knows it is 2 hops from LAN1, and

updates hosts on LAN3 with this information.
en

3. Router C broadcasts its requests and updates LAN3 that it is only 1 hop
Do

from LAN1.
4. Hosts on LAN3 use Router C to route messages to LAN1, rather than
Router B.
After hearing the routers’ broadcasts, a host selects the most cost-effective route
se

(the one with the fewest hops).

r
fe

Routes with 16 hops are

ea

considered inaccessible. You

can manually set up a router
to choose a route with a
greater cost if a situation
Pl
Re

requires it. For example, if

your server cannot service
clients appropriately, you
might want data routed to a
different server, even though
the cost is greater.

Lesson 11: IP Addressing, Subnetting, and Supernetting 269

 There are a few disadvantages to using distance vector routing with RIP, such as
the count-to-infinity problem. Convergence is extremely slow in a RIP environ-
ment, as displayed in the following figure.

convergence:
The synchronizing of all
routing tables in a RIP Count-To-Infinity Loop

ial
environment.
A count-to-infinity loop can occur among routers when a router or network goes
down. This can create an endless cycle of hop count recalculation. For example,
four routers connect five networks. In calculating the cost to Network 5, Router D
figures its cost to be one hop, Router C figures two hops, Router B figures three
hops, and so on. If Network 4 fails, Router C must recalculate its route table.
However, it still thinks that Network 5 can be reached, and uses information

er
advertised from Router B to calculate its table. According to Router B, Network
5 is three hops away, so Router C broadcasts that its cost to Network 5 is four
hops. Routers A and C receive the new information, update their tables to five
and three hops, respectively, and then broadcast their information. Router B
receives the new information and recalculates accordingly. This cycle continues to

at
infinity (the number specified by the implementation and configured by manage-
ment software—usually 16 hops).

y
op
eM
tC
No
c
en
Do

Figure 11-3: The count-to-infinity problem (slow convergence) associated with RIP.
se
r
fe
ea
Pl
Re

270
 ial
er
at
y
op
eM
tC
Figure 11-4: Count-to-infinity.
Let’s take a look at Figure 11-4 and see how the routers calculate hops. If Router
1’s connection to Network A breaks, Router 1 checks its information and deter-
mines that Router 2 has a one-hop connection to Network A. Realizing that it has
No
a direct connection to Router 2, Router 1 changes its route table to reflect a two-
c

hop route to Network A and starts forwarding its traffic to Network A through
Router 2. When Router 2 sees that Router 1 can now get to Network A in two
hops, it changes its own routing table entry to show that it has a three-hop path
en

to Network A, creating a routing loop (the count-to-infinity problem). The routing

Do

loop can continue indefinitely until the hop count maximum is exceeded, at which
time the route is marked as unreachable. Eventually the route is purged from the
route table. count-to-infinity:
A continuous cycle of hop
The following two algorithms help with count-to-infinity loops: count recalculation in which
• Split horizon , or best information—prevents a router from broadcasting a packet is never forwarded
se
r

to its destination.
route information received from its own network.
fe
ea

split horizon:
An algorithm that prevents a
router from broadcasting
internal network information.
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 271

 • Split horizon with poison reverse , or poison reverse—broadcasts 16 as the
route cost for transmissions on its own network.

poison reverse: Both algorithms yield the same results, but poison reverse takes less time. Also,
An algorithm ensuring that a poison reverse broadcasts two route tables to every network containing a router,
router broadcasts a route so updates are larger.
cost of 16 for all

ial
transmissions on its network.

Split Horizon
A work-around to the count-to-infinity problem is the split horizon method, where
a router does not include in its broadcasts any routes to the router from which it
discovered its own location. The following figure explains a situation in which

er
the split horizon method is used.

at
y
op
eM
tC
No
Figure 11-5: The split horizon method.
c

Router 1 calculated its hop count to Network C by hearing Router 2’s broadcast
that it is one hop from C. (Router 1’s hop count to Network C is two.) Because
the split horizon method is enabled, Router 1 broadcasts its routes to Network C
en

to all routers except Router 2. That way, if Router 2 goes down, hosts on Net-
Do

work B are not subjected to inaccurate information (created by the count-to-

infinity problem) when they recalculate their routing tables. The split horizon
method decreases RIP traffic but increases convergence time.
se
r

Poison Reverse
fe

Another work-around to the count-to-infinity problem in RIP networks is called

ea

poison reverse. Unlike in split horizon, routers using poison reverse broadcast
routes back to the router from which they calculated their own location, but
instead of giving a true hop count, they broadcast a hop count of 16, which
means they are inaccessible.
Pl
Re

272
 ial
er
at
Figure 11-6: The poison reverse method.

y
For example, Router 1 used information broadcast by Router 2 to calculate a hop
count of two to Network C. Router 1 broadcasts this information to hosts on Net-

op
eM
work A, but broadcasts a hop count of 16 to hosts on Network B. That way, if
Router 2 goes down, hosts on Network B will remove routing table entries to
Network C and will not try to get there through Router 1.

tC
In contrast to split horizon, poison reverse decreases convergence time but
increases RIP traffic.

Split horizon is enabled

RIP II
No
when poison reverse is
c

disabled and vice versa.

Because RIP is so simple (routers broadcast known routes and costs every thirty
seconds), RIP II was created. It enhances RIP (RIP I) by supporting the following
features.
en
Do

Feature Description
Next Hop Addressing Includes IP address information in route tables for every router in a
given path to avoid sending packets through extra routers.
Authentication Enables password authentication and the use of a key to authenticate
se

routing information to a router.

r

Subnet mask Supports more subnets and hosts on an internetwork by supporting

variable subnet mask lengths and including length information in
fe

routing information.
ea

Multicast packet Decreases the workload of non-RIP II hosts by speaking to only RIP II
routers. RIP II packets use 224.0.0.9 as their IP multicast address.
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 273

 Most hosts and routers support RIP I, so be sure that the RIP II mode you config-
For more information on RIP ure works with your current RIP configuration.
II, see RFC 1387 ″RIP
Version 2 Protocol Analysis.
″
Link-state Routing

ial
Unlike distance-vector routing, link-state routing provides for subsets of routing
You might also be interested
tables (instead of entire tables) to be broadcast by each router, which substantially
in RFCs 1388 and 1389 for reduces network traffic.
RIP II information.
Each link-state router broadcasts information about only those interfaces that are
directly connected to it. Each router builds a database based on link-state infor-

er
mation that it receives from other routers.
Novell’s NLSP protocol is an When a router’s status changes, the information that communicates that change is
example of a link-state distributed to all other routers by a link-state advertisement (LSA), which signals
routing protocol.
the other routers to update their databases accordingly. This process is called

at
flooding.

Link State Advertisement

y
(LSA):
A signal sent to other routers TASK 11B-1:

op
to update their databases.
eM
Understanding the Internet Protocol (IP)
1. Define autonomous system.

2.
tC
Which IGP uses RIP as its routing protocol?
No
c

3. How does the split horizon method relieve the count-to-infinity problem?
en
Do

4. Link-state routers broadcast to only those devices to which they are

directly connected.
se
r
fe
ea
Pl
Re

274
OSPF
On IP internetworks, link-state routing is usually accomplished by the Open
Shortest Path First (OSPF) protocol (RFCs 1245, 1246, 1850, and 2178). Each OSPF is a link-state
OSPF router uses the information in its database to build the shortest possible protocol. OSPF is an
path to destinations on the internetwork. implementation of link-state
routing.

ial
The following table compares the characteristics of RIP and OSPF.

Characteristic RIP OSPF

Size of metric 16. This means that a RIP network Limited only by the number of OSPF:
Open Shortest Path First. A
cannot be ″larger″ than 16 hops. bits in the metric field (64k).
link-state routing protocol
This maximum is further reduced Because OSPF does not suffer used on IP networks.

er
when costs other than 1 are used for from the count-to-infinity
certain routes. problem, it can be the basis for
much ″larger″ internetworks, and
system managers can assign
costs to optimize routing without

at
limiting the size of the network.
Maximum number of 15. This value is related to the 65,535. This value is related to
routers allowable metric size. the allowable metric size.

y
Variable-length subnets Only with RIP II. RIP treats subnets Supported by default. Because
as part of the internal structure of the OSPF treats the subnet mask as

op
network and assumes that all
eM part of the protocol information,
subnets are of equal length. With the restrictions that affect RIP do
RIP, all subnets must be contiguous, not apply.
connected, and hidden from remote

tC
networks.
Convergence Poison reverse or split horizon must LSAs provide rapid convergence
be used to counteract count-to- among tables; no count-to-infinity
infinity problem. RIP must calculate problem arises. OSPF passes Although OSPF uses less
bandwidth, it requires more
No
all routes before sending the along LSAs as soon as they are
memory and CPU resources.
information throughout the network. received, meaning that nodes can
c

adjust their routing tables at

practically the same time.
Traffic Entire routing table is broadcast Partial routing table (Hello packet)
en

every 30 seconds. is broadcast only to direct

Do

connections every 30 minutes.

se
r
fe
ea
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 275

 The OSPF Hello Packet
Adjacencies , or neighbor relationships, between routers that are connected to the
same segment are established and maintained by the OSPF Hello packet (RFC
2178). The Hello packet is a means of gathering routing information. It is used to
create a connection between two consecutive network routers, announce its
address and subnet mask, discover neighboring routers, establish the interval time

ial
Hello packet: between subsequent Hello packets, and discover the designated routers and
A means of gathering router backup designated routers (DR and BDR, respectively).
information when using
OSPF. DRs and BDRs are elected according to their priorities. The network router with
the highest priority becomes the DR, and that with the next highest priority
becomes the BDR. A DR or BDR is not replaced until and unless it goes offline.

er
For example, if a router with a higher priority than the DR is introduced to the
All network routers must network, it does not become the DR until the current DR becomes inactive. At
have a common Hello that time, the current BDR becomes the DR and the new, high priority router
interval, or they cannot
recognize each other.
becomes the BDR.
You can manually configure router priority, but be careful assigning a priority of

at
zero, since that prevents a router from becoming DR or BDR.

y
OSPF Route Discovery

op
eM
The OSPF protocol uses a two-step process to learn internetwork routes. First, it
establishes communication by using the Hello packet. Then it creates a database
of internetwork routes.

Route Database
tC
Once Hello packets are exchanged and the DR and BDR are established, each
No
router must create a database in which to store routing information and synchro-
c

nize it with those of the DR and BDR.

Prior to synchronization, routers exchange Data Description Packets (DDPs) with
en

the DR and the BDR. The packets contain a summary of routers’ databases. At
Do

this stage, the routers are in a two-way state . Once synchronization takes place,
each router is described as having an adjacency to one another and the relation-
ship changes to a full-neighbor state , as displayed in the following figure.
se
r
fe
ea
Pl
Re

276
 ial
er
at
y
op
eM
Figure 11-7: OSPF routers synchronize their databases.
Routers 1 and 2 send their summary packets (DDPs) to the designated router. In
return, the DR sends its DDP to R1 and R2 (step 1). The routers are in a two-

tC
way state. R1 and R2 compare the information in the DDP sent by the DR with
their current database entries. R1 sees that it has an outdated entry so it sends a
Link State Request (LSR) to the DR requesting the updated entry information.
The DR returns a Link State Update to R1 containing the requested information
No
(step 2). Finally, R1 incorporates the new information into its database and sends
c

a Link State Acknowledgment (LSA) to the DR (step 3). Having synchronized

their databases, the routers are now in a full-neighbor state.
The DR and BDR should have a full-neighbor relationship with each network
en

router. However, each network router should have a full-neighbor relationship

Do

with only the DR and the BDR.

se
r
fe
ea
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 277

 Route Selection and Maintenance
An OSPF router uses its database to see the entire internetwork. When creating a
route table, a router uses the OSPF algorithm to figure out how many hops it
takes to get to a particular network. Then it adds the number of hops to get the
OSPF will work properly only total cost of the route. For multiple routes to the same network, the route with the
if all routers agree on the lowest cost is added to the route table. The table information remains until a

ial
autonomous system’s link
change is reflected in a link-state advertisement. OSPF waits until a group of
state (the routers are
synchronized). changes have occurred, and then performs the algorithm again and rebuilds the
route table. By making several changes at once, OSPF does not run the algorithm
unnecessarily.
If an OSPF router detects a change in the network, it issues Link State Update
For more information on

er
(LSU) packets every thirty minutes (default setting). If no network change is
OSPF, visit the Cisco OSPF
Design Guide Web site at
detected, the router intermittently floods the internetwork with LSUs for every
http://www.cisco.com/warp/ entry in its route table. Each receiving router compares the LSUs with its data-
public/104/1.html. base and returns an acknowledgment to the sending router. Each router deletes
entries for which the aging timer has reached a value four times the Router Dead
interval.

at
An OSPF router re-synchronizes using DDPs only when its full state changes.
DDPs are also used when routers and the DR and BDR are out of sync. In this

y
case, adjacencies must be re-established.

op
eM
TASK 11B-2:

tC
Understanding the Open Shortest Path First (OSPF)
Protocol
1. What means does OSPF use to establish adjacencies between routers on
No
the same segment?
c
en

2. How are DRs and BDRs elected and which router becomes the DR?
Do

3. What happens if you assign a router a priority of zero?

se
r

4. What is the difference between a two-way state and a full neighbor

fe

state?
ea
Pl
Re

278
TOPIC 11C
Introduction to Subnetting
Each network segment requires its own unique network address, just as each

ial
street in your town requires its own unique street name. In the TCP/IP environ-
ment, network segments are called subnets if each network segment’s address
must be derived from a single IP network address. This situation often occurs if
your company’s network IP address is assigned by the InterNIC; if your network
has multiple physical segments, you must create multiple network addresses from
the single Internet IP address.

er
The process of subdividing a single network address to allow for unique network
addresses on each subnet is called subnetting . To subnet, use a custom subnet
mask to ″borrow″ bits from the host portion of the network address. (A subnet
mask distinguishes the host portion of the IP address from the network portion.)
Use the borrowed bits to create subnetwork addresses.

at
If your network does not connect to the Internet, then you may assign any unique
network address to each network segment and use the default subnet mask .

y
Although you will commonly hear this procedure described as subnetting, it is
more properly called segmenting , because there is no need to subdivide a single

op
network address.
eM
There are many reasons for subnetting or segmenting your network. They
include:

tC
• To reduce traffic by dividing a large network into smaller segments.
• To connect networks across geographical areas.
• To connect different topologies, such as Ethernet, Token Ring, and FDDI,
together via routers.
No
• To avoid physical limitations, such as maximum cable lengths or exceeding
c

the maximum number of computers on a segment.

• To increase security by limiting sensitive network traffic to one network.
en

• To help with troubleshooting by minimizing the impact of a subnet problem

Do

on other subnets.
If your network is connected to the Internet and has multiple segments, you must
subnet your assigned IP address. In order to implement subnet addresses on your
network, ask yourself some questions:
se

•
r

How many network addresses will I need? You will need one address per
subnet, including one for each connection to a wide area network.
fe

• How many host addresses will I need? This number determines the class of
ea

IP network address you can use for your network. You will need one host
address per computer, including one for each network card in a router or
printer.
Pl
Re

The answers to these questions will help you to define your network’s subnet
mask and network and host addresses.

Lesson 11: IP Addressing, Subnetting, and Supernetting 279

 TASK 11C-1:
Understanding subnetting concepts
1. Why would you subnet your network?

ial
2. Why would you segment your network?

er
at
3. If your network consists of three segments connected by two routers,
how many subnet (network) addresses will be necessary for proper con-
figuration of TCP/IP?

y
op
4.
eM
If your network consists of two segments, with 30 hosts each, connected
by a router, how many host addresses will you need?

Subnet Masks
tC
No
c

Every host on a TCP/IP network must have a subnet mask, even if the network
consists of only one segment or does not connect to the Internet. A subnet mask
is 32 bits long, the same length as an IP address. The subnet mask’s primary pur-
en

pose is to mask the host portion of the IP address in order to identify whether a
Do

transmitted packet’s destination is to a local or a remote host. However, you can

also customize the subnet mask and use it to subdivide a single assigned IP
address. The following table describes the two types of subnet masks and how
they accomplish their major functions.
se
r

Type of subnet
Subnet mask: mask Function Mechanism
A 32–bit address used to
fe

mask the host portion of the Default subnet mask Identifies local vs. remote Masks the host portion of the IP address for
ea

IP address. It is also used to transmissions. the sender and receiver of a packet, leaving
subdivide a single IP the network portion visible. TCP/IP compares
address into multiple the network address of the sender and
subnets. receiver to determine if the packet belongs on
Pl
Re

the local or remote network.

Custom subnet mask Also identifies local vs. Also masks the host portion of the IP address;
remote transmissions; borrows bits from the host portion of the IP
subdivides a single address to create additional network addresses
assigned network address. from a single assigned address.

280
TASK 11C-2:
Understanding subnet masking
1. Which portion of the IP address does the subnet mask hide?

ial
2. Why do you need a subnet mask if your company’s network doesn’t
connect to the Internet?

er
3. What other function does the subnet mask perform in the event your
network connects to the Internet?

at
y
op
Using Default Subnet Masks
eM
You can use the default subnet mask when your network does not require
subnetting. This situation usually occurs if your network is not connected to the

tC
Internet. The default subnet mask detects whether a packet’s destination is to the
local or a remote network. It is determined by the class of the network address.
The following table identifies the default subnet masks for Class A, B, and C IP
addresses.
No
c

Subnet
Network mask in
class Ranges Subnet mask in binary decimal
en

A 1.x.x.x - 126.x.x.x 11111111.00000000.00000000. 255.0.0.0

Do

00000000
B 128.0.x.x - 191.255.x.x 11111111.11111111.00000000. 255.255.0.0
00000000
C 192.0.0.x - 223.255.255.x 11111111.11111111.11111111. 255.255.255.
00000000 0
se
r

The 1s in the subnet mask enable the network portion of the IP address to be
fe

stripped from the host portion. The 1s enable the values in the network address to
ea

″pass through,″ while the 0s prevent the host portion from being seen.
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 281

 ial
er
at
Figure 11-8: Applying the default subnet mask.

y
op
eM
TASK 11C-3:

tC
Determining default subnet masks for different IP
Address Classes
1. When is the default subnet mask used?
No
c
en

2. If your IP address is 110.105.25.0, what is the default subnet mask?

Do

3. If your IP address is 198.105.25.0, what is the default subnet mask?

se
r

4. If your IP address is 135.64.4.10, what is the default subnet mask?

fe
ea

Using Custom Subnet Masks to Subdivide a Network

Pl
Re

Address
You must use a custom subnet mask to subdivide the single network address
assigned by the InterNIC. The subnet mask is customized when it ″borrows″ some
of the bits from the host address to identify the different subnet segments. All
computers within the same network must use the same subnet mask.

282
The number of subnets required for your physical network determines the number
of bits that must be borrowed from the host portion of the IP address. The subnet
mask requires that the equivalent number of borrowed bits be set to a value of 1.
The 1s in the subnet portion of the mask enable subnet addresses to ″pass
through.″ The custom subnet mask determines if the destination of a packet trans-
mitted on the network is local or remote. Packets transmitted to the Internet
appear as if they come from the single assigned network address, (essentially, a

ial
router) regardless of the subnet address, because the Internet applies only the
default subnet mask.
To calculate the number of available subnet addresses, use the formula 2 x - 2,
where x equals the number of borrowed bits for subnetting. Two of the available
addresses are subtracted because a value of all 1s or 0s in the subnet address is

er
not permitted.
The following example shows a custom subnet mask for a Class B address where
the entire third byte is used for subnet addressing.
Let’s examine the Class B IP address 145.98.20.5, with the subnet mask of 255.

at
255.255.0. The binary equivalent to the subnet mask would be 11111111.
11111111.11111111.00000000. The binary equivalent to the IP address would be
10010001.01100010.00010100.00000101. This would enable you to have a total
of 2 8 -2 = 254 subnets and 2 8 -2 = 254 hosts. The network address (including

y
the subnet address in the third byte) is 145.98.20.0 (10010001.01100010.

op
eM
00010100.00000000. The host address is --.--.--.5 (00000101).
Using the third byte to represent subnet addresses reduces the number of avail-
able host addresses per subnet to 254. The eight bits that are borrowed enable
you to have up to 254 subnets on this network.
However, suppose you are assigned a Class B network address from the
InterNIC, but need more than 254 host addresses for each subnet in your
network. If you borrow all eight bits from the third byte, you will not have tC
No
enough host addresses. You will need to borrow only enough bits from the third
byte to identify all subnets and leave the remaining bits for host addresses. In the
c

following example, only four bits have been borrowed from the third byte.
Let’s examine how this example would work when using the subnet mask 255.
en

255.240.0. The 145.98.20.5 binary value remains the same as in the previous If you want to learn more
Do

example. However, the subnet mask binary equivalent is now 11111111.11111111. about subnetting, take a
11110000.00000000. This would enable you to have a total of 2 4 -2 = 14 subnets TCP/IP course.
and 2 12 -2 = 4,094 hosts. The network address (including the subnet address in
the third byte) would be 145.98.16.0 (10010001.01100010.00010000.00000000);
the host address would be 0.0.4.5 (00000000.00000000.00000100.00000101).
se
r
fe
ea
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 283

 TASK 11C-4:
Using custom subnet masks
Objective: To determine custom subnet masks for different network
scenarios.

ial
1. What is the purpose of a custom subnet mask?

2. If you borrow three bits from the host portion of a Class B address, how

er
many subnets will you be able to assign unique addresses to?

3. How many host addresses will still be available after subnetting?

at
4. If your IP address is 145.8.17.96 and the subnet mask is 255.255.255.0,

y
what is the subnet address?

op
eM
tC
Configuring a Default Gateway
The default gateway address is the address of the router on the LAN through
which clients access another physical or logical network. If your LAN connects
No
to the Internet, this router is the point at which the two networks connect.
c

Each client that needs to access other networks (including the Internet) needs to
have an IP gateway address configured in their TCP/IP property settings.
en

Let’s use the following simulation to configure an NT Server as an IP router.

Do

These are the steps you would need to take to configure your NT server to act as
a router to connect your LAN to another subnetwork, a WAN, or to the Internet.

Suggested time:
APPLY YOUR KNOWLEDGE 11-1
se
r

20 minutes
Configuring a Windows NT server as an IP router
fe

In this activity, you’ll use the simulation to configure an NT server as an IP

ea

router, which could then be used as a default gateway to connect to another net-
work (such as the Internet).

1. Load the Activities component of the Network+ Certification CD-ROM.

Pl
Re

2. Click on the Network Protocol link, and then on the Subnetting And Routing
link.

3. Point to the Configure A Windows NT Server As An IP Router link. Read

the scenarios, and then click on the link.

284
4. Complete the steps in the Activity as prompted in the lower-right corner of
the screen.

5. When you finish the Activity, exit the Network+ Certification CD-ROM.

TOPIC 11D

ial
Supernetting
Supernetting , also known as Classless Inter-Domain Routing (CIDR) , is a
scheme devised to prevent the depletion of network addresses. Supernetting is

er
achieved by borrowing bits from the network ID and masking them as the host
ID. Supernetting supports address aggregation , where a single routing table entry Supernetting:
can represent the address space of possibly thousands of traditional classful routes Bits are borrowed form the
(routes that use Class A, Class B, and Class C addressing). This results in more network ID and masked as
the host ID. Used to help
efficient routing because it enables a single routing table entry to specify how to

at
prevent depletion of network
route traffic to many individual network addresses. addresses.
Supernetting eliminates the traditional concept of Class A, Class B, and Class C

y
network addresses and replaces them with the generalized concept of the network
prefix. Routers use the network prefix , rather than the first three bits of the IP
Classless Inter-Domain

op
eM
address, to determine the dividing point between the network number and the
host number. As a result, supernetting supports the deployment of arbitrarily sized
Routing (CIDR):
See supernetting.
networks rather than the standard 8-bit, 16-bit, or 24-bit network numbers associ-
ated with classful addressing.

tC
In the supernetting model, each piece of routing information is advertised with a
bit mask , or prefix length . The prefix length is a way of specifying the number
of leftmost contiguous bits in the network portion of each routing table entry. For
example, a network with 18 bits of network number and 14 bits of host number
No
would be advertised with an 18-bit prefix length. The IP address advertised with
c

the 18-bit prefix could, in fact, be a former Class A, Class B, or Class C IP

address. In such a classless environment, prefixes are viewed as contiguous
address blocks. Routers that support supernetting do not make assumptions based
en

on the first three bits of the address but rather rely on the prefix-length informa-
Do

tion provided with the route.

For example, assume your organization needs a network ID to accommodate
1,000 hosts. Rather than allocating a Class B network ID, the InterNIC allocates
a range of 4 Class C network IDs starting with 221.79.168.0 and ending with
221.79.171.0. Each Class C network ID accommodates 254 hosts for a total of
se
r

1,016 host IDs. Before supernetting, routers on the Internet would require the
following routing table for the four Class C addresses:
fe
ea

IP address Subnet mask Default gateway

221.79.168.0 255.255.255.0 221.79.168.1
221.79.169.0 255.255.255.0 221.79.168.1
Pl
Re

221.79.170.0 255.255.255.0 221.79.168.1

221.79.171.0 255.255.255.0 221.79.168.1

Lesson 11: IP Addressing, Subnetting, and Supernetting 285

 To supernet the four Class C addresses to accommodate 1,000 new hosts, you
would borrow 2 bits from the network number. In binary, the subnet mask would
change from 11111111.11111111.11111111.00000000 to 11111111.11111111.
11111100.00000000. This would make the new subnet mask 255.255.252.0. After
supernetting, routers on the Internet would require only the following entry in
their routing tables:

ial
IP address Subnet mask Default gateway
221.79.168.0 255.255.252.0 221.79.168.1

In supernetting, ANDing the destination IP address and the subnet mask of the
routing entry determine the destination packet. If a match for the network ID is

er
found, the route is used.

TASK 11D-1:

at
Supernetting IP Addresses

y
1. What is the purpose of supernetting?

op
eM
2. How does supernetting support the deployment of arbitrarily sized net-
works?

tC
No
c
en
Do

Summary
In this lesson, you’ve learned how to get an IP address assigned to your
organization, and how to implement it on your network. You also learned
how to make that IP address go farther by using subnetting and
se
r

supernetting.
fe
ea
Pl
Re

286
LESSON 11 REVIEW
11A Why are IP addresses divided into classes?

ial
11B Describe IP’s two routing methods.

er
at
11CList some advantages of subnetting a network.

y
op
eM
What is the main purpose of a subnet mask?
tC
No
c

11D Describe a supernet.

en
Do
se
r
fe
ea
Pl
Re

Lesson 11: IP Addressing, Subnetting, and Supernetting 287

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Accessing and Using a TCP/ LESSON
IP Network
12

ial
Data Files:
none
Overview
Lesson Time:

er
This lesson covers the use of a TCP/IP network. You’ll identify configura- 2 hours
tion parameters for workstations, and examine the use of TCP/IP utilities for
testing, validating, and troubleshooting IP connectivity.

at
Objectives

y
To learn how to access and use a TCP/IP network, you will:

op
12A Configure your workstation for TCP/IP.
eM
This topic covers the configuration of workstation parameters to connect
to an IP network. Parameters include IP address, DNS, default gateway,

tC
IP proxy, WINS, DHCP, host name, and Internet domain name.

12B Identify TCP/IP Suite Utilities.

You’ll be examining the use of ARP, Telnet, NBTSTAT, TRACERT,
No
NETSTAT, IPCONFIG, WinIPCFG, FTP, and PING.
c

12C Use TCP/IP Suite Utilities to test, validate, and troubleshoot network
connectivity.
en

In this topic, you’ll use the utilities from the previous topic to check out
Do

your IP network.
se
r
fe
ea
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 289

 TOPIC 12A
Configuring Clients for TCP/IP Networks
Let’s examine how to go about configuring Windows 95/98 and Windows NT

ial
workstations to use TCP/IP networks. You can configure it manually or via a
DHCP server. DHCP (Dynamic Host Configuration Protocol) is used to automati-
DHCP: cally assign or lease IP addresses to hosts.
Dynamic Host Configuration
Protocol. Used to To manually configure TCP/IP on a workstation, you’ll need to configure the fol-
automatically assign or lease lowing:
IP addresses to hosts.
•

er
IP address and subnet mask
• DNS (if you’re using DNS) to specify the DNS name
• Default gateway (if the computer is part of an internetwork)
DNS:
Domain Name System. • Any additional gateways used on the internetwork

at
Resolves host names to IP • WINS (if you’re using the WINS Service)
addresses.
• Computer name

y
These are all configured through the Network Control Panel. Add the TCP/IP pro-
tocol to the system, then access the TCP/IP Protocol properties dialog box.

op
default gateway:
The router or system that
eM
receives any packets that are
destined for a network that IP Address
isn’t included in the routing

tC
table. A unique IP address is required for every workstation and server on the network.
DHCP automatically assigns a different IP address to each workstation that
requests an address. When you are manually assigning IP addresses, you’ll need
to keep a worksheet or database of assigned IP addresses so you don’t inadvert-
No
ently assign the same address to multiple systems. Be sure to include all devices
c

to which IP addresses have been assigned, including workstations, servers, print-

ers, and routers.
The appropriate subnet mask must also be assigned. If the wrong subnet mask is
en

applied to the IP address, the system won’t be able to communicate with other IP
Do

devices on the network.

In the Network Control Panel, TCP/IP Properties dialog box, display the IP tab.
You can specify whether to obtain an IP address automatically (use this when
using DHCP to assign IP addresses) or to specify an IP address (use this when
se

you are manually assigning IP addresses).

r
fe

DNS
ea

In Windows 95/98, select the DNS Configuration tab, then select the Enable DNS
By default, you use the radio button. In Windows NT, just select the DNS tab.
Pl

computer name specified

Re

when you created your You’ll need to specify the computer name in the Host field. You can find your
system; however, you can computer name by right-clicking on the Network Neighborhood, choosing Proper-
specify a different name if ties, and then displaying the Identification tab. You’ll also need to enter your
you want to. domain name in the Domain text box (for example, global.com).

290
Next, you’ll need to add the IP address of the DNS server in the DNS Service
Search Order box. Finally, close the TCP/IP Properties box and restart your
workstation.
If you have secondary (backup) DNS servers, you’ll need to add those additional
server IP addresses. For each secondary DNS server in the domain, you’ll need to
add its IP address to the DNS Service Search Order box. The secondary DNS

ial
server is only used if the first DNS server fails to respond within a specific
timeout period.
If your domain is divided into subdomains, you’ll need to add those domain suf-
fixes (without the leading period) to the Domain Suffix Search Order list. When a
request is made without a domain qualification, the request is repeated throughout
the main domain suffix and through each of the additional domain suffixes in an

er
attempt to resolve the name to its IP address.

Default Gateway

at
If your system is part of an internetwork, you’ll need to specify the default
gateway. Packets destined for an IP address that isn’t on your local subnet or in
the route table are forwarded automatically to the default gateway. This is a

y
router or system connected to multiple networks that can determine how to get
the packet where it needs to go.

op
eM
You can have multiple gateways so that if the default gateway is unavailable, you
still have a chance to get your packet where it needs to go.

tC
Gateways are configured in the TCP/IP Properties box on the Gateway tab. Enter
the gateway device IP address, then click the Add button to add it to the Installed
Gateways list.
Let’s complete a simulation in which you’ll configure a default gateway.
No
c

TASK 12A-1:
en

Configuring a default gateway

Do

Objective: To complete the simulation in which you configure a default

gateway.
Setup: You’ll need the Network+ Certification CD-ROM to complete
se
r

this task.

1. Load the Activites component of the Network+ Certification CD-ROM.

fe
ea

2. Click on the Network Protocols link, and then click on the Subnetting
And Routing link.
Pl

3. Point to the Configure A Default Gateway link. Read the scenario, and
Re

then click on the link.

4. Complete the steps in the Activity as prompted in the lower-right corner of

the screen.

5. When you finish the Activity, exit the Network+ Certification CD-ROM.

Lesson 12: Accessing and Using a TCP/IP Network 291

 WINS
WINS (the Windows Internet Naming Service) is a program that runs on NT
servers. It associates a computer’s host name with its physical IP address,
enabling computers to find each other on the network. It is similar to how DNS
resolves Internet names. WINS is used to automatically resolve NetBIOS
addresses.

ial
You can configure Windows clients to use WINS. Each system with WINS
enabled registers with the WINS server at startup, thus creating a database of IP
addresses and NetBIOS names. This is configured in the TCP/IP Properties of the
client workstation.
If your system needs to be a WINS client, in the Network Control Panel on the

er
TCP/IP Properties dialog box, display the WINS Address or WINS Configuration
tab (the name changes between Windows 95, Windows 98, and Windows NT).
Add the IP address of the WINS server (or servers) that will be used to resolve
addresses for this workstation.

at
In Windows 95/98, you can also select to use DHCP for WINS Resolution. When
you select this option, the DHCP server is used to resolve the addresses for
WINS.

y
op
eM
tC
No
c
en
Do
se
r
fe

Figure 12-1: TCP/IP services: WINS.

ea
Pl
Re

292
TASK 12A-2:
Examining TCP/IP properties on your workstation
1. Open the Network Control Panel on your workstation. You can do this
through Control Panel, or you can right-click on Network Neighborhood and

ial
choose Properties.

2. Display the TCP/IP Properties. You can do this by selecting TCP/IP and
clicking the Properties button, or by double-clicking on TCP/IP.

3. Display the DNS tab. Is Enable DNS selected? Why or why not?

er
4. In the Host field, observe the computer name. It was assigned when your
computer was created, and can be seen on the Identification tab of the Net-

at
work Properties dialog box.

5. Display the IP Address tab. Is Obtain An IP Address Automatically

y
selected? If so, what does this mean?

op
eM
6. Close the Network Control Panel.

CHECK YOUR SKILLS 12-1 tC Suggested time:

No
30 minutes
c

Installing and configuring TCP/IP

In this activity, you’ll use the NextSim module on the Network+ Certification
CD-ROM to simulate installing and configuring TCP/IP manually and to use
en

DHCP.
Do

1. Load the Activities component of the Network+ Certification CD-ROM.

2. Click on the Network Protocols link. Then, click on the TCP/IP Fundamen-
tals link.
se
r

3. Point to the Install And Configure TCP/IP Manually link. Read the scenario,
and then click on the link.
fe
ea

4. Complete the steps in the Activity as prompted in the lower-right corner of

the screen.
Pl

5. When complete, click Exit to return to the Activities list.

Re

6. Do the Install And Configure TCP/IP To Use DHCP activity.

7. When you have completed both Activities, exit the Network+ Certification
CD-ROM interface.

Lesson 12: Accessing and Using a TCP/IP Network 293

 TOPIC 12B
TCP/IP Suite Utilities
There are many utilities available for using and troubleshooting TCP/IP network

ial
problems. Most of these are included with the TCP/IP protocol stack and are pub-
lic domain applications. The utilities we will be looking at are:
• ARP, used to display and modify the local ARP cache;
• Telnet, used for to remotely connect to the server via terminal emulation for
administration and troubleshooting purposes;

er
• NBTSTAT, used to check NetBIOS state over TCP/IP connections;
• TRACERT, used to check the route to a remote system;
• NETSTAT, used to show current TCP/IP connection statistics;
• IPCONFIG, used on Windows NT systems to display the current IP configu-

at
ration;
• ifconfig, used on UNIX systems to display the current IP configuration;
• WINIPCFG, used on Windows 95/98 systems to display the current IP con-

y
figuration;

op
•
eM
FTP, used to transfer files between systems;
• PING, used to verify that two computers can communicate (provides infor-
mation on host name, IP address as well as verifying that you can connect to

tC
the remote system); and
• NSLOOKUP, used to query Internet domain name servers.
No
ARP
c

The Address Resolution Protocol (ARP) is responsible for translating a host’s

ARP: software address to a hardware address (the node address that is set on the net-
en

Address Resolution Protocol. work interface card).

Do

The Internet layer protocol

that translates software
addresses to hardware
addresses.
se
r
fe
ea
Pl
Re

Figure 12-2: The ARP protocol.

294
To translate a host’s IP address to a hardware address, each host uses ARP. Every
host maintains a table in RAM for this purpose called the ARP table. The ARP
table is very similar to a host table, except that that ARP table doesn’t need to be
manually updated like the host table does. The ARP protocol updates entries in
the ARP table automatically. The ARP table is volatile, and entries usually persist
for only a few seconds.

ial
If your system’s ARP table doesn’t contain an entry for a particular host, a broad-
cast is sent to all hosts on the network requesting the address. If the destination RFC 826 covers the ARP
host is up and supports ARP, it sends its hardware address to your system. ARP protocol in detail.
then adds the address to the ARP table of your system. This process of updating
the ARP table is transparent to the users. On most implementations of TCP/IP,
you can examine your ARP table, or even make manual entries to the table, if a

er
destination doesn’t support ARP (which is unlikely).
Entries in the ARP table are purged at regular intervals to prevent the table from
becoming too huge and unmanageable. ARP -a is used to view the current ARP
cache. Enter ARP by itself to display the help for the command. You can use
other options to display only ARP information for a specific address.

at
There are several options for the ARP command. They include:

y
Option Description
Remember that the ARP
-a or -g Displays current ARP entries by interrogating the current protocol

op
options -a and -g are the
eM
data. If inet_addr is specified, the IP and physical addresses for only same.
the specified computer are displayed. If more than one network
interface uses ARP, entries for each ARP table are displayed.
Inet_addr Specifies an Internet address.

tC
-N if_addr Displays the ARP entries for the network interface specified by if_
addr.
-d Deletes the host specified by inet_addr.
No
-s Adds the host and associates the Internet address inet_addr with the
physical address eth_addr. The physical address is given as 6
c

hexadecimal bytes separated by hyphens. The entry is permanent.

Eth_addr Specifies a physical address.
en

If_addr If present, this specifies the Internet address of the interface whose
Do

address translation table should be modified.

RARP
se
r

RARP (Reverse Address Resolution Protocol) was adapted from the ARP proto-
col, and provides reverse functionality. It determines a software address from a RARP:
fe

hardware address. A diskless workstation uses this protocol during bootup to Reverse Address Resolution
ea

determine its IP address. Protocol. The Internet layer

protocol that translates
hardware addresses to
software addresses.
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 295

 TASK 12B-1:
Examining ARP options
1. Open a DOS window on your workstation.

ial
2. Enter ARP. It will display a summary of the command and list the
options used with the command. Record the options and a brief descrip-
tion of each.

er
at
y
op
eM
3. tC
Enter ARP -a to list the addresses in your cached ARP table. If there are no
entries in your table it’s not a problem—they’ve just expired!
No
c

4. Which options have the same function?

en

5. Close the DOS window.

Do

Telnet
Telnet is a terminal emulation protocol that enables a user at one site to simulate
se
r

telnet: a session on a remote host. It does this by translating keystrokes from the user’s
A terminal emulation terminal to instructions recognized by the remote host, then carrying the output
fe

protocol in the Process/ back to the user’s terminal and displaying it in a format native to the remote host.
ea

Application layer of the DOD This service is transparent (it gives users the impression that their terminals are
model that enables a user directly attached to the remote host). You can connect to a host that is running a
connect to a host computer telnet daemon (program), such as a UNIX host, or an IBM mainframe running
over the network and run a
TCP/IP. You can’t log on to Windows NT or NetWare servers via telnet.
Pl

terminal session.
Re

As defined in RFC 854, telnet uses the following defaults:

• Port 23, however, you can specify a different port if the host to which you
are connecting is configured to use a different port.
To telnet into an SMTP
server, you would use port • 25 lines in the buffer, but can be configured for up to 399 lines.
25.

296
• VT100 as the default terminal emulation, but some versions allow you to
configure your system with VT220, VT320, or IBM 5350 terminal emulation
support.
Windows includes a basic telnet utility. It is installed when you install TCP/IP on
your Windows system. It includes VT100, VT52, and TTY terminal emulation,
but includes no telnet daemon.

ial
er
at
y
op
eM
Figure 12-3: Telnet.

TASK 12B-2:
Connecting to a host using telnet
tC
No
c

1. Choose Start→Run, and then enter telnet . This will start the telnet appli-
cation on your workstation.
en

2. Choose Connect→Remote System. This is how you begin to establish a

Do

remote connection for telnetting into another system.

3. In the Connect dialog box, enter the IP address of the UNIX server, and
then click Connect.
se

4. If nothing appears, press [Enter] a few times to bring up a login prompt.

r

5. Log in as the user specified by your instructor. Most systems won’t allow
fe

you to telnet in as the root user for security reasons.

ea

6. Enter a command (such as ls -a to display the directory contents).

7. Log out and close the telnet application window.

Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 297

 FTP
The File Transfer Protocol (FTP) enables the transfer of files between a user’s
ftp: workstation and a remote host. Using FTP, a user can access the directory struc-
File Transfer Protocol. A ture on a remote system, change directories, search and rename files and
protocol in the Process/ directories, and download and upload files.
Application layer of the DOD

ial
model that enables the In order to access the directory structure on the TCP/IP host server, an FTP utility
transfer of files between a needs to be installed on the workstation. This could be the FTP program included
workstation and a remote in some releases of Windows, in the Novell LAN WorkPlace or WorkGroup prod-
host. ucts, or other utilities, many of which are available for download. The system to
which you are connecting must be running the FTP daemon in order for you to
connect to it using FTP.

er
You’ll also need a user name and password to connect to the FTP server. Many
FTP uses port 20 for data FTP servers allow limited access through an anonymous connection. To use this
and port 21 for commands. option, log in as the user anonymous, and enter your email address for the
password.

at
FTP is defined in RFC 959. It uses two TCP port connections: port 20 is used for
data and port 21 is used for commands.

y
FTP option Is used to
-v Prevent showing remote server command responses

op
-n
eM Suppress auto-login at initial connection
-i Disable interactive prompting when transferring multiple files
-d Debug, displaying all commands passed between the FTP client

tC
and server
-g Disable wildcard character support (disables filename globbing)
-s: filename Run all of the FTP commands contained in the filename file
-a Allow use of any local interface during data connection binding
No
-w: windowsize Override the default transfer buffer size
c
en

FTP Commands
Do

FTP commands must be entered in lowercase. There are both DOS and UNIX
commands included with the utility. To list all of the commands, at the FTP>
prompt, enter help . The Microsoft FTP utility includes the commands shown in
the following figure.
se
r
fe
ea
Pl
Re

Figure 12-4: The FTP commands.

298
TASK 12B-3:
Examining FTP commands
Objective: Use help to find out what the FTP commands enable you to
do.

ial
1. Open a DOS window.

2. Enter ftp .

3. Enter help and examine the list of commands available.

er
4. Enter help followed by one of the commands for additional help on what
the command is used for.

5. What are get, put, ls, open, and bye used for?

at
y
op
eM
tC
6. If you have access to an FTP server, log in (using either a valid user name
and password, or as an anonymous user) and try out some of the com-
mands; then try uploading and downloading a file.

7. Exit FTP using the bye command, and then close the DOS window.
No
c

NBTSTAT
en

NBTSTAT is a Windows 95/98/NT utility used to view and manage NetBIOS

Do

name cache information. It displays statistics about NetBIOS over TCP/IP NBTSTAT:
(NetBT) connections. Enter NBTSTAT at a DOS prompt to see a brief description A Windows utility used to
of the command and the command syntax. view and manage NetBIOS
name cache information.
In NetBIOS, names are 16 hexadecimal characters in length and are stored in a
flat-file database structure. You can set the first 15 characters in the name. The
se
r

last character indicates the resource type. Resources include:

• Workstation service (00) NetBT is a Microsoft addition
fe

to the TCP/IP protocol, so

•
ea

Remote access server (06) Microsoft created the

• File server service (20) NBTSTAT utility to
troubleshoot NetBT
• MS Exchange Message Transfer Agent (87) problems.
Pl
Re

Notice that the NBTSTAT

results show you what
services are running on the
remote system you’re
requesting information about.

Lesson 12: Accessing and Using a TCP/IP Network 299

 ial
er
Figure 12-5: NBTSTAT options.

at
y
op
eM
tC
No
c
en
Do

Figure 12-6: NBTSTAT results.

TASK 12B-4:
se
r

Examining NBTSTAT
fe

1. In a DOS window, enter NBTSTAT to display the usage message.

ea

2. Enter nbtstat -n to display the NetBIOS Local Name Table.

3. Enter nbtstat -r to display the NetBIOS Name Resolution and Registration

Pl
Re

Statistics.

4. Enter nbtstat -a NTSRVR to display information about the adapter in the

Windows NT classroom server.

5. If you want, try out more of the options for the nbtstat command.

300
TRACERT
You can use TRACERT to see what route data uses to get to its destination.
ICMP protocol Time Exceeded messages are used to determine the route. Packets TRACERT:
are sent out and each time a packet is sent the Time-to-live value is reduced A utility used to see the
before being forwarded. This allows the time-to-live to count how many hops it route data uses to get to its
is to the destination. destination.

ial
er
at
y
op
eM
Figure 12-7: TRACERT results.
Using this command as a troubleshooting tool, you can see how far the packets
tC
No
are getting when users can’t reach their destination. If a network firewall is con-
c

figured not to allow TRACERT or PING through, you might not be able to trace
the route all the way to the end; it might appear to end at the firewall.
en

If you get the message Destination Unreachable, a router isn’t able to figure out
Do

how to get to the next destination. Even though it doesn’t tell you what is wrong,
it alerts you to the router where the problem is occurring.
If you’re having trouble resolving host names when using TRACERT, use the -d
option to prevent TRACERT from trying to resolve host names. It also speeds up
response time since it isn’t spending time resolving host names.
se
r

The default number of hops TRACERT will attempt to reach is 30. Using the -h
option, you can specify more or less hops for it to check.
fe

With loose source routing, you specify the destination router and your local com-
ea

puter using the -j option. (TRACERT -j router local_computer). It lets you trace
the round trip route rather than the default with is just to get to the destination.
If many of your responses on the TRACERT are timing out, you can increase the
Pl
Re

number of milliseconds to wait before continuing. If, after increasing the value,
destinations are then reachable, you probably have a bandwidth issue to resolve.
This is specified using the -w option.
In summary, the TRACERT options are:

Lesson 12: Accessing and Using a TCP/IP Network 301

 Option Description
-d Don’t resolve addresses to host names.
-h max_hops Maximum number of hops to search for a target.
-j router local_system Loose source route along host list (router, computer).
-w timeout Wait the specified number of milliseconds for each reply.

ial
ROUTE
The ROUTE command is used to maintain network routing tables. You can also

er
use it to display and configure network routing tables. The syntax of this com-
mand is:
ROUTE [-f] [command [destination] [MASK netmask]
[gateway] [METRIC metric]]

at
The -f option clears the routing tables of all gateway entries. If this is used in
conjunction with one of the following commands, the tables are cleared prior to
running the command.

y
The command options used with ROUTE are:

op
Command
eM Used to
PRINT Print a route
ADD Add a route

tC
DELETE Delete a route
CHANGE Modify an existing route
No
The destination parameter specifies the destination host. All symbolic names used
c

for destination are looked up in the network database file NETWORKS. You can
use wildcards with the PRINT or DELETE commands.
The MASK parameter specifies that the next parameter is the subnet mask value
en

to be associated with this route entry. If no netmask is entered, it defaults to 255.

Do

255.255.255.
The gateway parameter specifies the gateway for the route entry. The symbolic
names for gateway are looked up in the host name database file HOSTS. This
parameter isn’t required when using the PRINT or DELETE commands. If you
do use it with PRINT or DELETE, you can use wildcards.
se
r

The METRIC parameter specifies that the next parameter is the cost-metric for
fe

this destination.
ea
Pl
Re

302
 ial
Figure 12-8: The ROUTE PRINT results.

metric:
TASK 12B-5: The cost associated with a
given route, usually

er
measured in hops, or the
Examining TRACERT and ROUTE commands number of routers that must
be used to complete the
1. In a DOS window, enter TRACERT . Read through the usage message is route.
that’s displayed.

at
2. Enter TRACERT www.microsoft.com and observe the results. Each gate-
way that the ICMP packets are sent through are recorded on your screen.

y
3. Did your packet make it all the way to the destination?

op
4.
eM
If necessary, press [Ctrl]C to end the TRACERT.

5. Enter ROUTE . A usage message is displayed; however, it is too long to fit

on the screen, so to see the beginning of it, you would need to press the

tC
Pause key to halt the display temporarily, then press Enter to continue.

6. Enter ROUTE PRINT to print out your current route table.

7. Close the DOS window.

No
c

NETSTAT
en

NETSTAT displays protocol statistics and current TCP/IP network connections.

Do

The following options can be used with the NETSTAT command. NETSTAT:
A Windows and UNIX utility
Option Used to used to display protocol
statistics and current TCP/IP
-a Show all connections and listening ports. Lets you know who the system network connections.
is communicating with and the current status.
se
r

-e Show Ethernet statistics. Lets you monitor communication traffic

amounts.
fe

-n List addresses and ports in numerical order. Lets you see the addresses
ea

rather than having them resolved to names.

-s Lists statistics for each protocol. Lets you display information for TCP,
UDP, ICMP, and IP protocols.
Pl
Re

-p UDP or -p TCP Display connections for the specified protocol. Lets you display
information for just one protocol.
-r Displays the routing table before the connection information. Lets you
see the network address, net mask, gateway address and interface, and
how many hops to the host.
interval (in seconds) Repeat the connection information at the specified interval

Lesson 12: Accessing and Using a TCP/IP Network 303

 TASK 12B-6:
Examining the use of NETSTAT
1. Open a DOS window.

ial
2. Enter netstat /? to display help information.

3. Record the available options and their uses.

er
at
y
4. Try out each of the options.

op
eM
5. Try the -a option with a 5 second update. Press [Ctrl]C to stop displaying
information.

tC
6. Close the DOS window.

Displaying Current Client TCP/IP Configuration

No
Information
c

There are times when you need to gather information about the computer you are
sitting at. You can find the IP address, subnet mask, and other information using
en

the utilities listed in the following table.

Do

Platform Utility
Windows 95/98 WINIPCFG
Windows NT IPCONFIG
se

UNIX ifconfig
r
fe
ea
Pl
Re

304
 ial
er
Figure 12-9: WINIPCFG results.
On a Windows 95/98 workstation, WINIPCFG displays the network card driver,
adapter address, IP address, subnet mask, and default gateway. The More Info WINIPCFG:

at
button displays additional information about the IP configuration. A Windows 95/98 utility that
displays the IP configuration
information.

y
op
eM
Figure 12-10: IPCONFIG results.

tion about the IP configuration, use the /all parameter with the IPCONFIG tC
On a Windows NT workstation, IPCONFIG displays IP configuration information.
This needs to be run from the DOS command line. To display additional informa- IPCONFIG:
A Windows NT utility that
No
command. displays the IP configuration
c

information.
en
Do
se
r
fe
ea

Figure 12-11: ifconfig results.

On a UNIX workstation, ifconfig displays the status of the currently active net-
work interface devices. Using options, you can dynamically change the status of
Pl

ifconfig:
Re

the interfaces and to change the IP address. A UNIX utility that displays
the IP configuration
information.

Lesson 12: Accessing and Using a TCP/IP Network 305

 PING
You can use the PING command to verify that a computer can be reached on the
PING: network. It checks the host name, IP address, and that the remote system can be
A utility for checking IP reached.
connectivity.
Ping uses the ICMP ECHO_REQUEST datagram to check connections between

ial
hosts by sending echo packets, then listens for reply packets.
Enter ping at the command prompt to list the options for this command. Some of
the options include setting the packet size, changing the TTL value, and specify-
ing how many times to ping the host.

er
at
y
Figure 12-12: PING results.

op
eM
TASK 12B-7:

tC
Checking IP address information
Objective: You’ll find out what your IP address is on your workstation,
then use PING to try to get a response from your workstation
No
and other workstations.
c

1. Using the appropriate TCP/IP configuration utility (IPCONFIG, WINIPCFG,

or ifconfig), find out what your IP address is.
en

2. Open a DOS window. Ping your workstation IP address.

Do

3. Ping your classroom servers’ IP addresses.

4. What other information can you find about your TCP/IP configuration
using the appropriate utility?
se
r
fe
ea
Pl
Re

5. Close the DOS window.

306
NSLOOKUP
Another command that you might find useful is NSLOOKUP. It is used to query
Internet domain name servers. The command has interactive and non-interactive
modes. This command is available on UNIX and NT servers.
Interactive mode enables you to query names servers for information about hosts

ial
and domains or to print a list of hosts in a domain. Interactive mode is used
when no arguments are listed; the default name server is used. It is also used
when the first argument is a hyphen and the second argument specifies a name
server.
The non-interactive mode prints only the name and requested details for a host or
domain. This mode is used when the name or IP address of the host is listed as

er
the first argument. The second argument is optional in this case.
For more information about this command, enter man nslookup on a UNIX sys-
tem or enter nslookup at the DOS prompt on an NT Server.

at
TOPIC 12C

y
Using TCP/IP Suite Utilities to Check Network

op
Connectivity
eM
Now that you’ve been introduced to the functions of some of the TCP/IP suite

tC
utilities, let’s examine how you can use those utilities to check network connec-
tivity and use them for troubleshooting network connectivity problems.
No
Troubleshooting with ARP
c

ARP can be used to help troubleshoot duplicate IP address problems and when a
workstation can’t connect to a specific host. Say you can ping the host and the
offending workstation from your system, but when you try to ping the host from
en

the offending workstation, you get Destination Unreachable error messages. Use
Do

the ARP -a command to display the current entries in the ARP table. If you see
that the MAC address listed is not the MAC address for the NIC in the host, you
can delete that entry from the ARP table by issuing ARP -d ip_address . Then,
add the correct entry by entering ARP -a ip_address mac_address . Remember to
check that the user can now access the host before closing this trouble ticket.
se
r
fe

Troubleshooting with Telnet

ea

Telnet is usually used to connect to another system and issue commands as

though you were sitting at the system. However, you can also use this utility to
help you troubleshoot problems. If upper-level OSI protocols or applications
Pl
Re

aren’t working, you can sometimes telnet into a system and send commands
directly to a server process.

Lesson 12: Accessing and Using a TCP/IP Network 307

 Troubleshooting with NBTSTAT
If you have Microsoft Windows computers on your network, NBTSTAT can be
very helpful in figuring out problems on those systems. Windows systems use
NetBIOS naming and NBTSTAT was developed specifically as a diagnostic tool
for use with NetBIOS. It provides additional information that isn’t available with
other TCP/IP utilities to examine NetBIOS information.

ial
Troubleshooting with PING and TRACERT
If you have a client who is unable to access a server, you might try using PING
and TRACERT to troubleshoot this problem. PING can be used to determine if

er
the server is running and TRACERT can be used to determine how far down the
line the problem is occurring that is preventing the user from accessing the
server.
If the server is running, you should receive a response right away when you ping

at
the server. Try issuing the PING command from another system to see if you can
reach the server. Then, have the client who is experiencing the problem try
pinging the server. They’ll probably get the message Destination Host

y
Unreachable. Then have the client issue a TRACRT command to the server to see
how far the system gets before receiving the Destination Unreachable error

op
eM
message. Using the IP address of the last successful connection, you’ll know
where to begin troubleshooting the problem, and hopefully even which device is
causing the problem.

Troubleshooting with NETSTAT

tC
If a user is having trouble using a TCP/IP-based program, such as an SMTP mes-
No
saging system or an FTP server, you might consider checking this problem out
c

using NETSTAT -a. For example, if you find that users can’t send or receive
messages, from the email server, enter netstat -a to find out if the system is lis-
tening on the expected port. Then, you can check to see if the messaging system
is up, or if it needs to be restarted.
en
Do

The -a option is used to check active connections. A good TCP connection usu-
ally has no bytes in the send or receive queues.
You can also check Ethernet NIC error counts. If the error count is up, check to
see if this indicates a problem (or it might just be a very busy network with lots
of collisions). You can also check routing tables for network routing problems
se
r

through NETSTAT.
fe
ea

Troubleshooting with IPCONFIG, WINIPCFG, and

ifconfig
Pl
Re

The IP configuration utilities IPCONFIG, WINIPCFG, and ifconfig are useful for
showing your workstation’s IP address. This is especially useful if you are getting
the address dynamically assigned by a DHCP server; otherwise, how would you
know what your address is?

308
Renewing and Releasing IP Addresses
If your DHCP server is configured to lease the address to you for a specific
amount of time, but the system doesn’t release the address, you can use these
utilities to force it to release the address. The method for doing it with each of
these utilities is listed in the following table. Also, if you want to keep the IP
address for awhile longer, you can renew the address using these utilities.

ial
Method for releasing IP Method for renewing IP
Utility addresses addresses
IPCONFIG ipconfig /release ipconfig /renew
WinIPCfg winipcfg /release_all or winipcfg winipcfg /renew_all or winipcfg /renew

er
/release adapter or winipcfg, then adapter or winipcfg, then click More
click More Info, and click the Info, and click the Renew button
Release button

at
Troubleshooting using FTP

y
Many vendors have FTP sites from which you can download patches and other
support information. This is the main purpose FTP will serve in your trouble-

op
eM
shooting toolkit. You can help automate the FTP download process by creating a
text file containing the FTP commands to be issued, then calling this on a Win-
dows system with the -s option.
For example, if you want to connect to the ftp.novell.com site, then download the

tC
current patch for NetWare 4.11 to a directory on your workstation, you could cre-
ate the file NW411DL.TXT with the appropriate FTP commands. When you want
to download the file, enter ftp -s NW411DL.TXT and it will use those FTP
commands.
No
c

TASK 12C-1:
en
Do

Determining which TCP/IP utility to use

Objective: Read through each of the scenarios listed, and then determine
which TCP/IP utility or utilities you would need use to deter-
mine the problem. Tell why you chose to use those utilities.
se
r

1. You’re working the helpdesk and get a call that a user can’t access the
UNIX host at 150.150.32.157. You are on the same subnet as the user
fe

and the UNIX host and try to ping the UNIX host. You can successfully
ea

do so. You can also ping the user’s workstation. When you ask the user
to enter ping 150.150.32.157, all they get is a series of Destination
Unreachable messages. What should you do?
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 309

 2. The results of the listing displayed from step 1 shows the UNIX host,
but the MAC address listed isn’t the MAC address for the NIC in that
system. What should you do?

ial
3. You’re a network administrator and have been receiving complaints that
users aren’t able to post files to or download files from the FTP server.
What might you do to determine the status of the server?

er
4. A Windows 95 user logs in and gets a message that their IP address is
already in use. This user receives their IP address through a DHCP
server. How can the user get a valid IP address so they can do their

at
work?

y
op
5.
eM
A client calls from the California office saying that they aren’t able to
connect to the server ICANY in New York. This server is on a routed IP
network. This is the second client from California who has called with

tC
this problem. No users from other sites have called. What should you
do?
No
c
en
Do

6. You need to download the latest patches for several operating systems
and software applications. Which troubleshooting tool would be best
suited to this task?
se
r
fe
ea
Pl
Re

310
APPLY YOUR KNOWLEDGE 12-1 Suggested time:
15 minutes
Identifying TCP/IP utility output
In this activity, you’ll examine the output displayed in each step below, then iden-
tify the utility and any options used to create that output.

ial
1.

er
2.

at
y
op
eM
3.

tC
No
c
en
Do
se
r
fe
ea

4.
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 311

312
Re
fe
6.
5.

Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
7.

ial
er
8.

at
y
9.

op
eM
tC
No
c
en

10.
Do
se
r
fe
ea
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 313

 Suggested time:
CHECK YOUR SKILLS 12-2
10 minutes
Reviewing TCP/IP utilities
In this activity, you’ll work through some sample questions that deal with the
information presented in this lesson.

ial
1. Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

er
3. Select all items for the TCP/IP Suite: Utilities topic, and uncheck the Test
Time check box, and then click OK.

4. Work through the sample questions and score the test.

at
5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

y
6. Exit the test.

op
eM
tC
Summary
In this lesson, you configured your workstation for use with a TCP/IP
network. You then used several TCP/IP suite utilities to display information
about your TCP/IP connection. Finally, you identified which TCP/IP suite
No
utility would be used to troubleshoot network problems.
c
en

LESSON 12 REVIEW
Do

12A What do you need to configure on your workstation to establish a

TCP/IP network connection?
se
r
fe
ea
Pl
Re

314
12B Identify the purpose of each of the following TCP/IP suite utilities: ARP,
Telnet, NBSTAT, TRACERT, NETSTAT, IPCONFIG, WINIPCFG, FTP,
and PING.

ial
er
at
12CWhat utility would you use to find a problem in the path between two

y
IP addresses?

op
eM
What utilitity would you use to make sure an IP address is reachable?

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 12: Accessing and Using a TCP/IP Network 315

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Remote Access LESSON
13

ial
Overview Data Files:
Many users now work at home or away from the office frequently. This les- Network+ Certification
son introduces you to the protocols used to establish remote-access CD-ROM

er
connections to the network. You’ll also examine how to install and configure
dial-up networking on a Windows-based workstation. Lesson Time:
1 hour, 30 minutes

Objectives

at
To learn more about remote-access connections, you will:

y
13A Examine serial protocols.

op
eM
This topic takes a look at the SLIP and PPP remote access protocols.
You’ll also take a brief look at the Point-to-Point Tunneling protocol.

13B Examine remote connection options.

tC
You’ll compare the attributes, advantages, and disadvantages of ISDN and
POTS communication options. We’ll also briefly examine some of the
other options available.
No
13C Configure modems for dial-up networking.
c

To finish this lesson, you’ll check out the modem configuration param-
eters used for dial-up networking. You’ll also pull all of this lesson’s
information together as you identify the requirements for a remote
en

connection.
Do
se
r
fe
ea
Pl
Re

Lesson 13: Remote Access 317

 TOPIC 13A
Remote Access Protocols
Remote users usually connect to your network either directly or via an ISP over

ial
analog phone lines. These analog connections are implemented via an analog
modem. The connection can use the Serial Line Internet Protocol (SLIP) or Point-
To-Point (PPP) connection. These are the two standard protocols for connecting
remote devices over dial-up connections.

er
SLIP
SLIP allows IP bytestreams to be sent over serial lines. The IP datagram is
SLIP: appended with a SLIP END character to distinguish separate datagrams. This is
Serial Line Internet Protocol. also known as a packet-framing protocol because each packet is framed with

at
An older method for sending these characters. Both ends of the transmission need to convert the data back to
IP bytestreams over serial IP datagrams. SLIP is defined in RFC 1055.
lines (such as modem/phone
connections). The following table lists some positive and negative features of the SLIP

y
protocol.

op
Advantages
eM Disadvantages
Can be used on many types of media No packet addressing
Has been a standard in UNIX-based systems since No packet ID information

tC
1984
No error detection or correction
No compression features
No
c

PPP
en

PPP is the Internet standard for sending IP datagram packets over serial point-to-
Do

PPP: point links. It can be used in synchronous and asynchronous connections. IP, IPX,
Point-to-Point Protocol. The and AppleTalk network layer protocols can be encapsulated using NCP (network
Internet standard for sending control protocol). PPP is defined in RFC 1661. It can dynamically configure and
IP datagram packets over test remote network connections.
serial links (such as modem/
se

phone connections). This is the protocol that should be used by Windows clients when connecting to a
r

Windows NT network or to the Internet. PPP has replaced SLIP in most

situations.
fe
ea

A PPP server needs to

provide this service.
Pl
Re

318
PAP and CHAP
To successfully log in using PPP, you need to enable PAP (Password Authentica-
tion Protocol) and CHAP (Challenge Handshake Authentication Protocol). When
the link between the dial-in client and the server is established, a PPP node
requests that the data-link peer (in this case the client) authenticate by using
either PAP or CHAP. If the peer doesn’t agree to provide the requested authenti-

ial
cation data, the peer is rejected, the link isn’t created, and the call is terminated.
If the peer agrees to send the authentication data, the link is established and
authentication begins.
The protocol that the server and the peer agreed on is then used to authenticate
the remote peer. If authentication is unsuccessful, the call is terminated.

er
PAP was the first authentication protocol used by PPP to identify peers. The ID
and password are both sent as clear text. The ID and password pair submitted by
the remote peer are compared to a local list of authorized IDs and passwords. If a
match is found, authentication is successful; if no match is found, the link and
physical connection are terminated.

at
CHAP was developed to prevent having to send the password as clear text. A
secret is maintained by both peers. One peer sends a challenge sequence. Using

y
the secret, this sequence is modified; the sequence is then returned to the peer
that sent the challenge. This peer then validates the sequence by using its secret

op
eM
on the original challenge; the peer then compares the results to the sequence it
received from the other peer. Authentication then proceeds as it does when using
PAP.

Comparing PPP and SLIP

The benefits of PPP over SLIP are that PPP provides user authentication, testing tC
No
the usability of the link (line monitoring), and the ability to dynamically config-
c
ure the address. Before a SLIP connection can be established, the IP address must
be configured each time. PPP can dynamically configure the IP address when the
connection is started.
en

PPP includes data compression features, error detection and correction, and
Do

encryption to allow more secure transmissions than with SLIP.

SLIP requires more modem script configuration to establish the connection. It
needs an 8-bit, no-parity, hardware-flow-control connection. Also, SLIP has more
overhead, causing it to be slower. However, some ISPs still use SLIP because it
has been in use longer than PPP.
se
r
fe
ea
Pl
Re

Lesson 13: Remote Access 319

 PPTP
If you need more security than allowed by PPP connections, you should consider
PPTP: using the Point-To-Point Tunneling protocol. It uses multiprotocol virtual private
Point-to-Point Tunneling networks (VPNs) which allow secure communication over standard Internet
Protocol. A protocol that connections. This allows you to create VPNs over the Internet.
connects clients and servers

ial
by using Remote Access Rather than invest in expensive dedicated connections between sites, you can use
Services (RAS). PPTP can be VPNs to communicate securely between sites. PPTP can also be configured for a
used to create a virtual network card so that only PPTP packets are sent and received.
private connection between
computers over the Internet. PPP connections are used to transport the packets. The PPTP data is encrypted
It is an extension of PPP. then encapsulated within PPP packets. These packets can be TCP/IP, IPX, or
NetBEUI packets. The encryption key is negotiated between the remote access

er
server and the client when the PPP connection is created.

The server needs to be able Windows 95/98 and Windows NT systems can use PPTP. The protocol needs to
to support the PPTP service. be installed on both ends of the transmission (the client and the server).

at
TASK 13A-1:

y
Comparing remote connection protocols

op
1.
eM
Compare SLIP and PPP connections.

tC
No
2. Which protocol would you need to use if you are using a VPN?
c

3. Which protocol is capable of transmitting multiple protocols over a

en
Do

serial link?

4. When would you use PPTP?

se
r
fe
ea
Pl
Re

320
TOPIC 13B
Examining Remote Connection Options
There are many connection options for connecting systems together including

ial
analog phone lines, dedicated connections, ISDN, T1 (through T4) connections,
and Frame Relay to name some of the most common. For a client sitting at his or
her home system or remote office and needing to dial into your network, the most
practical of these connections are the analog and ISDN connections since they are
more widely available and less costly than the other options.
Let’s take a closer look at the features and requirements for these two options.

er
We’ll also examine the advantages and disadvantages of each of these options.

Analog Connections

at
Analog connections over standard phone lines are available just about anywhere
in the world. These connections are implemented via an analog modem. The con-
nection can use SLIP or PPP connections.

y
These connections run over the standard copper phone lines found in most

op
eM
homes. These are referred to as PSTN (Public Switched Telephone Network) or
POTS (Plain Old Telephone Service) connections. Since this medium was origi-
nally designed for human voice communication, analog connections were just
fine. However, as computers needed to communicate with one another, access to

tC
those lines were modified; modems were used to connect digital information into
a form that could be transmitted back and forth across the analog lines.
No
Equipment
c

Equipment needed for analog connections includes modems. Servers might also
use UARTs (Universal Asynchronous Receiver/Transmitters), routers, or both.
en

The modem establishes the serial link on the phone line. Modems are available at
Do

different baud rates, or speeds. The most common baud rate at the time of this
writing is 56 Kbps. You might still encounter some 28.8 and 33.6 Kbps modems.
When using 33.6 Kbps or 56 Kbps modems, be sure the phone line supports traf-
fic over 28.8 Kbps. Most modems have some type of compression to improve
line speed. The modem should use the V.34 standard.
se
r

Instead of a modem, you can use a router. This is more expensive than a modem,
but has the advantage of being able to connect an entire network rather than just
fe

Analog lines are sometimes

one system. referred to as the local loop.
ea

A UART in the server increases serial port performance on the server. Some
modems require use of a UART for proper performance.
Pl
Re

Cost
If you don’t already have an available analog line, you’ll need to have one
installed. If you want to connect to a system and still be able to talk on the phone
at the same time, you’ll need to have two installed.

Lesson 13: Remote Access 321

 If you connect to the server through an ISP, you might incur a startup fee for
connection software or for developing modem scripts. There will be a monthly
service charge from the phone company and from the ISP. Sometimes paying for
several months or a year in advance can lower these fees.

ial
Bandwidth
The bandwidth for this type of connection is based on the modem speed (28.8,
33.6, or 56 Kbps for example). These connections are widely available and are
inexpensive compared to other connection options. The FCC has limited the
speed of analog transmissions to 53 Kbps; however, it is unusual for many cir-
cuits to connect over 44 Kbps because of noise on the line. The bandwidth is

er
very narrow, connections are often poor, and the framing of SLIP and PPP pack-
ets causes a lot of overhead.

ISDN Connections

at
Integrated Services Digital Network (ISDN) connections are becoming more
widely available, mainly in major metropolitan areas. These are popular for

y
telecommuting and dial-up access. ISDN uses a single wire or fiber optic line to
carry voice, data, and video signals. For example, you can be speaking on the

op
eM
phone and accessing a remote computer at the same time over a single ISDN
line. It uses existing phone company switches and wiring, which are upgraded to
make a 64 Kbps end-to-end digital channel. Japan and North America use one

tC
standard; Europe uses another.
The Basic Rate Interface (BRI) is most commonly used in residential ISDN
connections. It’s composed of two bearer (B) channels at 64 Kbps each for a total
of 128 Kbps (used for voice and data) and one delta (D) channel at 16 Kbps
No
(used for controlling the B channels and signal transmission). The total bandwidth
c

is up to 144 Kbps.
The Primary Rate Interface (PRI) is most commonly used between a PBX (Pri-
vate Branch Exchange) at the customer’s site and the central office of the phone
en

company. It is composed of 23 B channels at 64 Kbps and one D channel at 64

Do

Kbps. The total bandwidth is up to 1,536 Kbps.

In Europe, the BRI is the same as above, but the PRI is composed of 30 B chan-
nels and one D channel. The total bandwidth is up to 1,984 Kbps.
se
r
fe
ea
Pl
Re

322
 ial
er
at
y
Figure 13-1: ISDN connections.

op
eM
Equipment

tC
A Network Terminal Interface (NT1) device is needed to connect your data or
telephone equipment to the ISDN line. It provides connection terminal equipment
(TE) and terminal adapter (TA) equipment to the local loop. This is a coding and
decoding device. It takes the place of a modem in an analog situation.
No
The TA replaces a modem. It is used to adapt ISDN BRI channels to RS-232 and
c

V.35 standards. It can be a standalone device or an interface card that is installed

in a computer, router, or PBX.
en

You might also need other ISDN interfaces and ISDN LAN topology bridges.
Do

There are two types of ISDN interfaces. The U interface is used to carry ISDN
signals over a single pair of wires over long distances such as from the central
office to you. The Subscriber/Termination (S/T) interface is used to carry signals
over two pairs of wires over a short distance such as from the wall jack to your
ISDN devices. When using the S/T interface, you’ll also need to use an NT1
se
r

device to convert between the U and ST interfaces. If your adapter connects

directly to the U interface, the NT1 device isn’t necessary.
fe
ea

Costs
ISDN installation costs vary. The costs include the charge by your ISP (if you
Pl
Re

connect to your network via an ISP) to use ISDN, the cost to install and config-
ure ISDN hardware and software, and the cost for technical support.
Continuing costs include local loop and bandwidth fees. A flat fee is usually
available. You might also choose a basic rate with a per-call charge incurred for
the length of time channels are in use. The local loop charge might be high if you
require intense use of the ISDN channels.

Lesson 13: Remote Access 323

 Bandwidth
BRI has two 64 Kbps channels. These can be bonded together to provide 128
Kbps bandwidth. This bonding occurs at the phone company’s central office. Fre-
quently six channels are bonded for a 384 Kbps bandwidth.
PRI has 23 64 Kbps B channels. These can be bonded for a 1,472 Kbps

ial
bandwidth.
The D channel is used to reassign channels for voice, fax, and data as required.

T-carrier Categories and DS Levels

er
T-carriers are the most common high-speed Digital Data Service lines in use. Of
the T-carriers, T-1 is the most often used. Data Signaling or Digital Service (DS)
lines are categorized into five levels. The T-carriers are categorized into four
categories. The following table describes these categories.

at
T-carrier
category Contains DS level Bandwidth
DS-0 doesn’t correspond to

y
any T-carrier. It’s the data A single T-1 channel DS-0 64 Kbps
rate of a single T1 channel, T1 24 channels; 1 T1 DS-1 1.54 Mbps

op
64 Kbps. T2
eM 96 channels; 4 T1s DS-2 6.31 Mbps
T3 672 channels; 28 T1s DS-3 44.74 Mbps
T4 4,032 channels; 168 T1s DS-4 274.18 Mbps

Digital Data Service tC

No
DDS or Digital Data Service lines are most often used in synchronous, direct
c

point-to-point WAN connections. You can also lease a dial-up or dedicated line
connection for private use. These DDS connections don’t use modems; they use
CSU/DSU equipment. They usually use PPP and HDLC protocols. T-carrier lines
en

are the most common type of DDS line. Other DDS connection types include:
Do
se
r
fe
ea
Pl
Re

324
DDS connection type Description
56K Uses a partial T1 line. Switched 56 uses it on a dial-up basis.
56 K line:
Dedicated 56 is always up. Switched 56 is often used as a
Switched or dedicated 56 K
redundant service in case the dedicated line goes down. connection using a partial T1
X.25 Uses switched virtual circuits to dynamically connect remote line.
terminals to mainframes. Connections are maintained until the user

ial
shuts it down.
Frame Relay Uses fast-packet technology that takes advantage of fiber optic
transmission media. It is a connection-oriented, non-reliable X.25:
connection. Switched virtual circuits used
to dynamically connect
ATM Uses fixed-length packets called cells in a packet-switching network. remote terminals to
Uses error-correction bits. Data transfer rates are 622 Mbps or

er
mainframes.
better.
SMDS Uses the same technology as ATM, but uses no error checking. Data
rates of 1.544 Mbps to 45 Mbps can be attained.
SONET Uses fiber optic media. Data rates are defined as OC or optical Frame Relay:
carrier levels. OC1 is 51.84 Mbps, OC3 is 155.52 Mbps, and OC24 A fast-packet technology that

at
is 622 Mbps. uses fiber optic lines.

y
ATM:

op
TASK 13B-1:
eM Asynchronous Transmission
Method. High-speed packet-
switched technology using
Discussing digital connections 53-byte cells and error-
checking.

tC
1. What equipment is needed for ISDN connections?

SMDS:
No
Switched Multimegabit Data
c

2. Compare European with American and Japanese ISDN connections. Service. Similar technology
to ATM, but uses no error
correction.
en
Do

3. Compare the use of analog lines with ISDN lines. SONET:

Synchronous Optical
Network. Uses fiber optic
media for high-speed
communications.
se
r
fe

4. What are the functions of the B and D channels used in ISDN?

ea
Pl
Re

5. How many channels are each of the T-carriers composed of?

Lesson 13: Remote Access 325

 6. What equipment is used with Digital Data Service rather than use of
modems?

7. Compare ATM and SMDS services.

ial
TOPIC 13C

er
Dial-up Networking
You’ll need to configure remote access software on the remote computer that you
will use to dial into the remote access server. For remote node connections,
among the options you have are Windows NT 3.51 and 4.0 remote access ser-

at
vices, DOSDIAL, Windows Dialer (through Dial-Up Networking in Windows 95
and Windows NT), LAN Workplace 5.0, and Novell Mobile Services.

y
op
Modems
eM
For most workstations that will be remotely dialing in to the network, you’ll need
to install a modem. Modems can be internal or external devices. In a server envi-

tC
ronment, they tend to be external. Another popular solution in server
environments is to use an external rack mounted with multiple internal modems.
External modems have LEDs that can be useful in seeing whether the modem is
For a more detailed transmitting and receiving data. Internal modems don’t have this feature, although
No
discussion of modems, refer it is often simulated in software.
c

to http://www.zdnet.com/
products/stories/howtobuy/
Modems can also be hardware-based or software-based (often referred to as
0,7561,2252542,00.html. controller-less modems or Winmodems). Software modems are slower than hard-
ware modems because of the overhead in translating code. They leave the
en

Follow the links on the side

processing to the PC CPU rather than incorporating a controller chip on the card
Do

of the page to go directly to

a topic, or use the Next link as the hardware modems do. The main advantage of doing so is that they can be
at the bottom of each page to sold very inexpensively. A disadvantage is that these modems aren’t designed for
read through all of the NT, Linux, or other operating systems besides Windows 95/98.
related articles.
Modern modems include fax and voice capabilities. This allows you to use your
se

PC to send documents as faxes and to use your PC as a phone and answering

r

machine.
fe
ea

Asynchronous and Synchronous Modems

Asynchronous (or async for short) modems are the most common type of modem.
Pl
Re

Serial data transmission includes Start and Stop bits in the data stream to indicate
the beginning and end of each character.

326
The data flow is controlled by the slower of the two systems. This slower system
interrupts the transmission whenever the buffers are full and it needs time to
catch up. Since this method has great potential for errors, a parity bit is used for
parity check error control. This is why many PCs are configured with 8 data bits,
none for parity, and 1 stop bit (often referred to as 8-none-and-one).
Synchronous modems are less prone to errors. They use a timing mechanism to

ial
regulate transmissions between systems. Data is sent in frames that contain synch
characters before each frame. The frame only needs to be re-sent if the synch
character doesn’t arrive at the prescribed point.
Several synchronous protocols have been defined as listed in the following table.

Protocol Description

er
BSC Binary Synchronous Communications Also known as bi-sync. Replaced by
SDLC. Developed for use in
connecting IBM 360 mainframes and
IBM 3270 terminals. Frames include
header and trailer synch characters.

at
SDLC Synchronous Data Link Control Designates one system to be the
controlling system and the other to be
controlled. Replaced by HDLC.

y
HDLC High-level Data Link Control Bit oriented protocol that supports

op
eM variable length frames.

tC
ITU Modem Standards
Modem standards, or V dot modem standards, are defined by the ITU (Interna-
tional Telecommunications Union). Some standards have bis or terbo version
No
suffixes; these are French for second and third.
c

The most recent standard is V.90. This includes technology that enables receiving
data faster than 56 Kbps by bypassing modulation of the data. When sending
data, it must be modulated, resulting in a slower 33.6 Kbps data rate.
en
Do

The following table describes some of the most popular ITU standards.

ITU standard bps Notes

V.32 9,600 Synchronous is 9600; Asynchronous is
4800
se
r

V.32 bis 14,400 (14.4K) Synchronous and Asynchronous

V.34 28,800 (28.8 K)
fe

V.34 bis 33,600 (33.6 K)

ea

V.42 57,600 Specifies standards for error checking

V.42 bis Specifies standards for compression
V.90 56,000 (56K)
Pl
Re

Lesson 13: Remote Access 327

 MNP Modem Standards
The Microcom Networking Protocol (MNP) defines classes 1 to 5 for modem
standards. Each class offers different levels of error correction and detection.
Most modems use the MNP Class 5 protocol, which includes data compression.
This basically provides a data transmission rate of twice the speed you would
normally achieve.

ial
AT Commands
The AT or Attention commands are the modem command set used on most
modems. Hayes originally developed it for its line of modems. Since other com-

er
panies wanted to be ″Hayes compatible,″ they developed their modems to use this
command set. The AT command set is used by almost all PC modems.
The most basic command is AT, which alerts the modem that you want to com-
municate with your modem. The following table lists some of the most common

at
commands.

Command Description

y
For more information on the
Extended AT Command Sets, AT Attention used at the start of modem command lines.
DT Dial using touch tone.

op
visit http://www.modems.
com/general/extendat.html. H0
eM Hang up or disconnect.
This should go next to the
table. A Answer.
DP Dial using Pulsing (rotary dial).

tC
, Pause (each comma is roughly 3 seconds by default). Often used when you
need to dial an access code for an outside line so that you wait for the dial
tone. For example, 9,,,7162407300.
*70 Disable call waiting.
No
c

Installing and Configuring a Modem

en
Do

Before you start with the configuration, make sure your modem is connected to
the workstation computer and turned on. It is not necessary to have a phone cable
plugged into a telephone line for configuration purposes. Your modem is installed
through the Control Panel Modems utility. If no modems are currently installed,
the Install New Modem Wizard will automatically start; otherwise, click Add to
se
r

begin the installation. Follow the prompts, either allowing the wizard to auto-
detect your modem or you specifying the modem manually. Upon completion,
you can then configure the modem. If you have a Plug-and-Play modem, when
fe

you restart the system after installing it in a slot, the wizard will automatically
ea

start.
Pl
Re

328
 ial
er
at
y
op
eM
Figure 13-2: The connection page of the Control Panel Modems utility.
To configure the modem, use the Control Panel Modems utility. Select the
modem, then click Properties. The options that can be configured are listed in the
following table.

On the Properties tab:

Properties Description
tC Enabling the FIFO buffers in
your modem configuration
No
enables the UART on the
Port Specifies the port on which the modem is modem. This increases the
c

installed. speed of transmissions.

Speaker Volume Specifies the volume of phone tones including the
dial tone and dialing sounds.
en
Do

Maximum Speed Specifies the highest modem connection speed.

On the Connection tab:

Properties Description
se
r

Connection Preferences Used to configure the Data, Parity, and Stop bits.
fe

Call Preferences Specifies whether to wait for a dial tone before

dialing, to cancel the call if it doesn’t complete the
ea

connection (limit set in seconds), or to disconnect

the call if the system remains idle for too long
(limit set in minutes).
Pl
Re

Port Settings Accesses the Advanced Port Settings dialog box.*

Advanced Accesses the Advanced Connections Settings
dialog box.**

Lesson 13: Remote Access 329

 * If you need to configure your system to use a UART processor to increase the
speed of serial connections, it is configured here. You can also increase or
decrease the Receive Buffers and Transmit Buffers settings to correct connection
problems or get faster performance.
** This dialog box contains options to enable error control and flow control. You
can also specify the modulation type if you are using a non-standard modem

ial
(such as a Bell or HST). If you need other modem settings configured, you can
use the Extra Settings text box to enter the modem string to do so. You can also
enable logging of modem errors by checking the Append To Log check box. The
file C:\WINDOWS\MODEM.TXT is created. Commands sent to the modem and
modem settings sent by the modem are written to this file.

er
OPTIONAL TASK 13C-1:
Installing a modem
You don’t need a modem to

at
complete this task. 1. Open the Modems Control Panel utility. If no modems are installed, the
Install New Modem Wizard will start; if there is a modem already installed,

y
you’ll need to click Add to start the wizard.

op
2.
eM
Check Don’t Detect My Modem, and then click Next.

3. Verify that Standard Modem Types is selected in the Manufacturers list,

then select Standard 28800 bps Modem in the Models list. Click Next.

tC
4. From the port list, select COM2 and then click Next.

5. In the Location Information dialog box, enter your area code and, if
required, any access number required to access outside lines. This dialog
No
box might not be displayed depending on your system configuration. Notice
c

that either Pulse or Tone dialing can be selected in this dialog box. Click
Next.
en

6. Click Finish.
Do

IRQs and I/O Addresses

Let’s quickly review the default IRQs and I/O Addresses associated with the
se
r

COM ports. Remember that those devices with the same IRQ can’t be used at the
same time even though you can configure multiple devices with the same IRQ. In
fact, if Plug and Play is enabled, you can’t configure devices to share an IRQ at
fe

all. So, you couldn’t have COM1 and COM3 or COM2 and COM4 in use at
ea

once.

Port IRQ I/O Address

Pl
Re

COM1 04 03F8 - 03FF

COM2 03 02F8 - 02FF
COM3 04 03E8 - 03EF
COM4 03 02E8 - 02EF

330
The two Control Panel utilities on Windows 95/98 systems used to configure the
modem are Device Manager and Modem Configuration. Many modems also
include custom configuration utilities.

OPTIONAL TASK 13C-2:

ial
Configuring a modem
1. Select the modem you just installed, and then click Properties.

2. Display the Connection page and observe the Connection preferences

er
available. Options include Data bits, Parity, and Stop bits; and Call prefer-
ences, including how long to wait before disconnecting and waiting for a
dial tone before dialing.

3. Click Advanced. In the Advanced Connection Settings dialog box, you can

at
determine whether to use hardware or software flow control and you can
record a log file. Click Cancel to return to the previous screen.

y
4. Close all open Modem Property windows, and then close the Control
Panel. The Control Panel Modems utility is also used to remove modems.

op
eM
Select the modem to be removed, and then click Remove to remove it.

Configuring a Remote Access Connection Through Dial-

Up Networking
On a Windows 95 or Windows NT Workstation computer, you can use Dial-Up
Networking to create a connection that you will use to connect to the remote
tC
No
access server. Before you can do this, you must make sure that Dial-Up Network-
c

ing is installed.
1. You can find Dial-Up Networking in the Accessories program group.
en

2. If it has not been installed, go to the Control Panel and click on Add/
Do

Remove Programs.
3. Select the Windows Setup tab, and then select Communications.
4. Click Details and check the Dial-Up Networking option.
5. Click OK twice.
se
r

If you are using a Windows NT Workstation machine, Dial-Up Networking is

installed by default during the operating system installation.
fe

To configure a remote access connection on a Windows 95 Workstation computer:

ea

1. Select Dial-Up Networking from the Accessories program group.

2. Double-click on Make New Connection, and type a name for the connection
in the Type A Name For The Computer You Are Dialing text field. This
Pl
Re

name might describe the location into which you are dialing in to or you
might use your company’s name if you have only one dial-in connection to
your company’s network.
3. Verify that your modem is the modem listed in the Select A Modem drop-
down list box. If not, select your modem from the list.

Lesson 13: Remote Access 331

 4. Click Next. If you did not install a modem during the installation of the
operating system or at a later time, you will be prompted either to have the
system detect your modem or to pick a modem from the list. It is recom-
mended that you let the system detect your modem.
5. In the Make New Connection dialog box, enter your area code and telephone
number. Add a 9 and a comma before your telephone number if you need to

ial
dial 9 to reach an outside line.
6. Verify that your country is properly selected in the Country Code drop-down
list box.
7. Click Next.
8. Click Finish. The new connection is displayed in the Dial-Up Networking

er
program group.
The setup procedure for Windows NT Workstation computers is almost identical.
The information you provide is the same, although the steps are slightly different.
Start the installation of a new connection by selecting Dial-Up Networking from
the Accessories program group. A wizard then guides you through the connection

at
creation process.

y
Establishing a Remote Access Connection

op
eM
Make sure your modem is connected, turned on, and your telephone cable is con-
nected to the modem and your phone jack. In Dial-Up Networking, to establish
the connection between the workstation and the remote access server:

tC
1. Double-click on the connection you created.
2. In the User Name text field, enter the name of the user you want to connect
as.
3. In the Password text field, enter the user’s remote access password.
No
c

4. Verify that the phone number in the Phone Number text field is correct.
5. Click Connect. A dialog box displays telling you that the modem is dialing.
If you have a sound card installed, you might hear the modem dialing. After
en

a short while, a message tells you that the user name and password are being
Do

verified.
6. A message box appears that tells you that you are connected and the speed
and duration of your connection are displayed.
7. If you click the Details button, you can also see the server type and sup-
se

ported protocols for the connection.

r

8. You are now connected to the network through the remote access server, and
fe

you can browse and use the network resources.

ea

When you first browse a NetWare network, you are prompted to enter your NDS
user name and your NDS password, which is different from the remote access
password you enter when establishing the dial-up connection.
Pl
Re

To break the remote access connection, click Disconnect in the Connected To

message box.

332
OPTIONAL TASK 13C-3:
Configuring and testing a Dial-Up Networking
connection

ial
1. At the workstation, open My Computer, double-click on Dial-Up Net-
working, and begin to create a new Dial-Up Networking connection.

2. Name the new connection Recon 1 . (Note: Windows NT workstation users

will need to check the Send Plain Text Password and Non-Windows NT
options.)

er
3. Type the area code and phone number into the appropriate fields. (Note:
Windows NT workstations users will need to verify that PPP is selected, and
then accept the defaults in the remaining dialog boxes.)

4. Click Finish when the new connection has been created.

at
5. In Dial-Up Networking, double-click on Recon1 to begin testing the
connection. Type the name and password. Establish the remote access

y
connection by clicking Connect. Observe the messages in the Connecting
To message box that inform you of the current status of the connection.

op
eM
Once connected, the Connected To message box tells you the speed and
duration of the connection.

6. Break the remote access connection.

APPLY YOUR KNOWLEDGE 13-1 tC Suggested time:

No
20 minutes
Installing and testing a modem
c

In this activity, you’ll use the Network+ Certification CD-ROM to complete this
activity.
en

1. Load the Activities component of the Network+ Certification CD-ROM.

Do

2. Click on the Network Cables And Devices link; then, click on the Remote
Connectivity link.

3. Point to the Remote Connectivity link. Read the scenario, and then click on
se
r

the link.

4. Complete the steps in the Activity as prompted in the lower-right corner of

fe

the screen.
ea

5. When you finish the Activity, exit the Network+ Certification CD-ROM.
Pl
Re

Lesson 13: Remote Access 333

 Suggested time:
CHECK YOUR SKILLS 13-1
15 minutes
Reviewing remote connectivity
In this activity, you’ll work through some sample questions that deal with the
information presented in this lesson.

ial
1. Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

er
3. Select all items for the Remote Connectivity topic, uncheck the Test Time
check box, and then click OK.

4. Work through the sample questions and score the test.

at
5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

y
6. Exit the test.

op
eM
tC
Summary
In this lesson, you’ve examined several serial protocols, including SLIP,
PPP, and PPTP. You also compared the use of POTS and ISDN connections.
Finally, you examined how to install, configure, and test a modem, then
No
establish a remote connection.
c
en

LESSON 13 REVIEW
Do

13A List the serial protocols discussed in order from least secure to most
secure.
se
r

13B Describe characteristics of POTS and ISDN connections.

fe
ea
Pl
Re

334
13CIf Dial-up Networking isn’t installed on your workstation, how do you
install it?

ial
er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 13: Remote Access 335

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Network Security and Fault LESSON
Tolerance
14

ial
Data Files:
none
Overview
Lesson Time:

er
Network security is one of the most important topics in the networking 2 hours
industry today. It is up to you as the network administrator to help devise
and implement a network security policy that protects the data on your
network. It is also important that the data be always accessible by those that
have a legitimate need to access the data. This lesson looks at security mod-

at
els, password security, data encryption, the use of firewalls and proxy
servers, and ways to ensure high availability of that data.

y
op
Objectives
eM
To learn about network security and fault tolerance, you will:

tC
14A Examine security models.
Microsoft networking uses user and share level security models. Other
network operating systems use similar security methods to protect access
to information on your network. You’ll be examining the various models
No
available.
c

14B Examine password security.

One of the first lines of defense against network intruders is protecting
en

access and data with passwords. You’ll need a password to access the
Do

network and, sometimes, additional passwords to access specific files or

applications. This topic examines what defines a secure password and
identifies standard password practices and procedures.

14C Examine data encryption.

se
r

Another line of defense against intruders is to encrypt the data so that

even if files are accessed, the information contained in them can’t be read
fe

easily. We’ll be examining some data encryption techniques that you can
ea

employ on your network.

14D Examine the use of firewalls and proxy servers.

Pl
Re

If your computer wasn’t connected to any other computer, it would be

safe from all except for physical contact security breaches. However, that
would make the system less useful in today’s modern networked world.
Through the use of firewalls and proxy servers, you can help regulate
who can get to and out from the network and at what level they can
access the network.

Lesson 14: Network Security and Fault Tolerance 337

 14E Examine fault-tolerance features.
If you have all of the security measures in the world in place and a drive
dies, you still won’t have access to the information. This topic looks at
some of the system fault-tolerance features you can implement to main-
tain high availability of network resources. These features include
mirroring, duplexing, disk striping, and tape backups.

ial
14F Identify system backup policies and procedures.
Since the data on your network is only as good as your most recent
backup; in this topic, you’ll examine various backup choices available on
NetWare and NT networks.

er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

338
TOPIC 14A
Security Models
Security is the protection of data and system resources from access and manipula-

ial
tion by unauthorized users. Implementing security is the role of the network
administrator. Some simple measures of security are login names and passwords,
and hardware, such as a lock and key on your workstation.
Security is necessary to:
• Prevent access by malicious users, such as former employees who were let

er
go or other persons who wish to harm the company in some way.
• Ensure that people from outside of the company can’t access network
resources and data. Such people could include competitors and those
involved with industrial espionage.
• Accommodate management-level needs. Upper-level, mid-level, and other

at
workers need differing levels of access to different parts of the network. For
example, information about employee salaries and performance reviews
might be stored on the network. Not everyone on the network should be able

y
to access such files. You can enable security so that each of these groups of

op
eM
users has just the right amount of access to the resources they need.
• Ensure data security. NT provides user-level and share-level security to pro-
tect the network resources. NetWare provides login, file system, and NDS
security to protect network resources.

sider the following:

• tC
As you work to determine what to include in your security model, be sure to con-

Sensitivity of information stored on the network and who has access to this
No
sensitive information.
c

• Whether your LAN will be connected to the outside world (via public net-
work), and if so, if any information on your network should be accessible to
outside users.
en

• Establish password guidelines and any restrictions on passwords.

Do

• Determine host name naming guidelines.

• Determine who has access to which network resources.
Security management isn’t concerned only with protecting data, but is also
se

involved with more tangible functions:

r

• Equipment.
•
fe

Who has what hardware and software.

ea

• Inventory-control procedures.
• Maintaining lists.
• Reducing inside and outside hazards (for example, theft and destruction).
Pl
Re

Lesson 14: Network Security and Fault Tolerance 339

 Implementing Security on Your Network
Different network operating systems implement security in different ways. Win-
dows 95/98/NT uses share and user level security models, and NTFS permissions.
NetWare uses login, file system, NDS, and (with NetWare 5.0) printing security
features. UNIX implements security for users, groups and others by assigning file
system access to the resources to which the user or group has access (all network

ial
resources are treated as files in the UNIX system).
You can also secure your server to control who has access to the physical com-
puter and console.

er
Levels of NetWare Security
NetWare has several levels of security:
• Login security ensures that individuals can’t access network resources unless
they log in to the network by using a valid User object name, and if

at
required, a valid password. Requiring passwords is highly recommended. It
can also control how, when and from where a user can log in.
• File system security controls users’ access rights to the directories and files

y
NetWare 3.x doesn’t use stored on the network.
NDS, so that level of security

op
is not available for NetWare •
eM
NDS security controls users’ rights to portions of the NDS tree and, through
3.x networks. it, the file system.
• In NetWare 5.0, printing security controls users’ access rights to the printing
resources on the network.

tC
NetWare security is established through the Windows-based NetWare Administra-
tor utility. NetWare 4.x also includes a DOS-based utility, NETADMIN that can
be used. NetWare 3.x login security is established through the SYSCON utility
No
and FILER is used to establish file system security.
c
en
Do
se
r
fe
ea
Pl
Re

340
 ial
er
at
y
op
Figure 14-1: NetWare security.
eM
tC
Levels of Windows 95/98/NT Security
If you have File and Printer Sharing installed and enabled on your computer, you
can share folders and printers on your computer with other users. When you
No
share a folder, your computer is functioning as a file server. When you share a
printer, your computer is functioning as a print server.
c

Security for shared resources is partially determined by the method used to con-
trol other users’ access to the shares. Windows 95/98 offers two different types of
en

access control: share-level and user-level. Share-level access control, the default
Do

type, means that each share on a computer is accessed individually, and can be
protected by a password specific to each user, and are verified when the user ini-
tially logs on to the network. User-level security requires the presence of a
special computer, called a domain controller, which contains an access control list
of all users and their network rights. For example, Windows NT servers can con-
se

tain access-control lists and function as domain controllers.

r

You’ll need to have the Client for Microsoft Networks installed and configured to
fe

use share-level security. The file and printer sharing service for Microsoft Net-
ea

works must also be installed.

Share-level Access Control

Pl
Re

Share-level access control is easy to implement and maintain because users can
create shares and assign or change passwords as they work. As long as other
users know the passwords, they can access the shares. However, users must
remember the individual passwords for each share they wish to access, and the
passwords must not be distributed to unauthorized persons.

Lesson 14: Network Security and Fault Tolerance 341

 ial
er
Figure 14-2: Share-level access control.

at
User-level Access Control

y
User-level access control is more centralized than share-level access control and

op
eM
is therefore more secure. It is also more complex, because different users can
have different levels of access to a given share. The user’s access rights are vali-
dated by the domain controller at logon time, so the user needs to remember only
one password.

tC
Because of its security and one-step validation process, user-level access control
is the better choice for networks that have a domain controller available. Because
of its complexity, it’s best to assign a network administrator to plan, implement,
and maintain user-level access control.
No
c
en
Do
se
r
fe
ea

Figure 14-3: User-level access control.

Pl
Re

Changing the Access-control Type

Access-control type is a property of each specific computer. All the shares on a
computer will have the same type of access control.

342
If you change access-control type, from share-level to user-level or vice versa,
you’ll have to re-share all of the resources on that computer.
To implement user-level access control:
1. Open the Network Control Panel.
2. Select the Access Control tab.

ial
3. Select User-Level Access Control. Enter the network path to the computer
that contains the access-control list.
4. Click OK. You’ll need to restart your computer and re-share any resources
on it for the changes to take effect.
The following table shows the share-level permissions you can set.

er
Share permission Description For Windows
No Access Enables the specified user or group to make the NT
remote connection to the folder, but doesn’t
allow them access or the ability to view the

at
contents of the folder; this permission overrides
any other shared folder permissions the user is
granted individually or from membership in

y
another group.
Read Enables the specified user or group to make the NT, 95, 98

op
eM
remote connection to the folder, view the
contents of the folder, open the file and view
the contents, and run program files.
Change Gives the specified user or group the same NT

tC
abilities as the Read permission, plus the ability
to create and delete subfolders, add and delete
files, modify files, and change file attributes.
Full Control Gives the specified user or group the same NT, 95, 98
No
abilities as the Change permission, plus the
c

ability to change file permissions and take

ownership of files on NTFS volumes.
en
Do

Passwords in Share-level Security

One of the most basic elements of network security is the password. In many
networks, you’ll need a password as well as a user name to complete your initial
se

logon. You might also need a password for each shared resource that you access,
r

especially if you’re using share-level access control.

fe

You can password-protect a shared folder on your workstation by changing its

ea

Sharing properties. If you wish, you can assign different passwords for read-only
and for full access. After you’ve assigned the password, you must distribute it to
the people who will need to have access to that share.
Pl

Shared printers can also be password-protected. There is only one access level for
Re

shared printers.

Lesson 14: Network Security and Fault Tolerance 343

 TASK 14A-1:
INSTRUCTOR DEMONSTRATION—Sharing a folder by
using Windows Explorer

ial
1. Log in to the NTSRVR as Administrator.

2. Using Notepad, create a text file. Name the file NetPlus.txt and save it to
the C:\Temp folder.

3. Close Notepad.

er
4. Run Explorer.

5. Select and observe the Temp folder . The Temp folder is not
currently shared.

at
6. Choose File→Properties, select Sharing, and then select Shared As.

7. In the Share Name text box, accept the default name.

y
8. In the Comment text box, type Miscellaneous files.

op
eM
9. Observe the User Limit box. You can restrict the number of users who can
simultaneously connect to the shared resource.

tC
10. Click Permissions. The Access Through Share Permissions dialog box is
displayed. By default, Everyone has Full Control.

11. Click Add. The Add Users And Groups dialog box is displayed.
No
12. Click Show Users. You can assign share permissions by group or individual
c

users.

13. Select Domain Users, and then click Add to enable users in the domain to
en

access this share.

Do

14. Open the Type Of Access drop-down list. Read is the default access. You
can change the access to No Access, Change, or Full Control.

15. Close the Type Of Access drop-down list. We’ll leave the access set to
se

Read.
r

16. Click OK.

fe
ea

17. In the Name list box, verify that Everyone is selected. The group Everyone
automatically includes any user who can access the computer, including such
accounts as the Guest account. For security reasons, it is prudent to grant
share permissions only to those groups or users you specifically want to be
Pl
Re

able to access the resource. Click Remove.

18. Click OK to limit the remote access for your shared folder to any user in
your domain with Read access.

344
19. Click OK to close the Properties dialog box.

20. Observe the Temp folder icon . The icon indicates that the
folder is shared.

ial
Accessing Shared Resources
After a folder is shared, you can map a drive to it and access the shared resource.
Let’s check out how you would go about access the resource that was shared.

er
TASK 14A-2:
Accessing shared resources
Objective: To access the Temp folder your instructor shared.

at
Setup: Your instructor has shared the Temp folder on the NTSRVR
machine.

y
1. Log on to the network, accessing the NTSRVR.

op
2.
eM
In Explorer, choose Tools→Map Network Drive.

3. In the Path text box, enter \\ntsrvr\temp , and then click OK.

4. Observe the icon for the mapped drive. It looks like a mapped drive icon
rather than a shared folder icon.
tC
No
c

5. Double-click on the NetPlus.txt file in the shared folder to open and use
the file from the shared folder.
en

6. Close the file and close Explorer.

Do

UNIX Security
Everything on a UNIX system is treated as a file, so to implement access to files
se
r

and network resources, you need to give the users rights to the files that relate to
the directory, file, or network resource. This is done using the chmod command
to give or remove read, write and execute (rwx) permissions for user, group, and
fe

others. To implement a default permission for files or directories, use the umask
ea

command to set the default permissions.

Pl
Re

Lesson 14: Network Security and Fault Tolerance 345

 TASK 14A-3:
Discussing security models in the three main network
operating systems

ial
1. What are the levels of NetWare security in a NetWare 4.11 network?

2. What command is used to set permissions on a UNIX system?

er
3. What are the levels of Windows security?

at
User-level Security
A more secure option than share-level security is user-level security. However,

y
you can only use user-level security if you have a security provider. This could
be a Windows NT PDC (primary domain controller) which can authenticate the

op
eM
user. A user is validated using the PDC then matched with the assigned access
permissions.

Enabling User-level Security

tC
In order to establish user-level access control, you’ll need to make sure the Client
for Microsoft Networks is configured so that Log On To Windows NT Domain is
No
checked and the appropriate domain is listed. As in share-level security, file and
c

printer sharing needs to be enabled also. A common network communication pro-

tocol must be installed on all computers on the network.
en

In the Network Control Panel Properties, display the Access Control page. Select
Do

User-level Access Control, and then specify the domain through which users will
be authenticated. Alternatively, you can specify a NetWare server for authentica-
tion if the Microsoft Client For NetWare Networks and File and Printer Sharing
for NetWare Networks are installed. Accept the changes; then reboot your system
when prompted.
se
r

Sharing with User-level Security

fe
ea

User-level sharing starts out the same way as share-level sharing did: in Explorer,
select the folder; then right-click on it and select Sharing. Select the Shared As
option; then click Add and identify the users and groups who will be sharing the
folder. You can specify that they receive Read Only, Full Access, or Custom
Pl
Re

access to the shared folder.

Custom Access Rights allows you to choose from the following rights: Read
Files, Write to Files, Create Files and Folders, Delete Files, Change File
Attributes, List Files, Change Access Control.

346
TASK 14A-4:
Establishing user-level security access to a shared
folder
Objective: To use a Windows NT domain to authenticate and access a

ial
folder using user-level security access.

1. Display the Network Properties dialog box. Click File And Print
Sharing. Check I Want To Be Able To Give Others Access To My Files.
Click OK. Reboot when prompted, and then log in.

er
2. Display the Network Properties dialog box. Display the Access Control
tab.

3. Select User-level Access Control. This will enable you to specify the users
and groups who have access to each shared resource.

at
4. Verify that the Obtain List of Users and Groups From text box contains
NETDOMAIN . Click OK.

y
5. When prompted, reboot your system and then log in.

op
6.
eM
Open Explorer. Create a folder named your_name.

7. Right-click on the your_name folder. Choose Sharing. Select Shared As.

tC
8. Click Add; then select another student’s user name to allow them to
share the folder. Click on Custom. Click OK.

9. Check Read Files, Write To Files, Create Files And Folders, and List
No
Files. Click OK.
c

10. Grant Administrator Full Access Rights.

en

11. Click OK until you return to Explorer.

Do

12. Observe the folder icon. It now has a hand under the folder to indicate that
it has been shared.

TOPIC 14B
se
r
fe

Password Security
ea

Passwords electronically protect data. Most NOSs have a logon security system
that requires user validation and authentication by entering a unique user name
Pl

and password. Passwords should be changed frequently and whenever you sus-
Re

pect that unauthorized access to your data has been made. Develop a corporate
password policy and disseminate that policy to your user community. Such a
policy, like that enforced by the Windows NT Account Policy, could describe

Lesson 14: Network Security and Fault Tolerance 347

 minimum password lengths, frequency of password change, and suggestions for
password formats. Your password policy should include a directive that pass-
words should never be given to anyone for any reason. Hackers often pose as
computer support personnel and ask users for their passwords.

ial
Designing Password Guidelines
When you design your network, you should include guidelines for password cre-
ation and use. Most network operating systems include ways to specify password
requirements and settings. Areas that you should consider including in your pass-
word guidelines document might include:
•

er
Minimum password length
• How often the password must be changed
• Whether to allow users to ever reuse a password
• How many attempts a user has to login with an invalid user name or pass-

at
word before the account is locked
• Whether users set their own password or if an administrator assigns the pass-
word and then tells the user what it is

y
op
eM
Choosing a Good Password
A secure password is not a word you can find in a dictionary. For more secure
passwords, use long passwords that include numbers, letters, and possibly special

tC
characters such as hyphens or underscores. Don’t use family members’ names or
pets’ names, or other easily guessed passwords. Never post your password; any-
one walking by could see it and use it.
No
c

TASK 14B-1:
en

Discussing password security

Do

1. What are passwords?

se
r

2. How is a password policy implemented in an NT environment?

fe
ea
Pl
Re

348
3. What are some areas that a password guidelines document might cover?

4. List some examples of good passwords. List some examples of poor

ial
passwords.

er
TOPIC 14C

at
Data Encryption

y
Closely related to passwords is data encryption . Data that must pass through
unsecured data communications lines (like WAN links) or to outside agents (like

op
eM
other companies) should be secured. One way to do this is to encrypt the data.
Encryption is the process of converting the data into coded form in such a way
that only authorized parties can access the information. Only those with the nec-
essary password and decryption key can decode and read the data.

tC
encryption:
Many encryption schemes and methods are available. Algebraic, transposition, or The process of converting
substitution methods can be used to encrypt the data. Electronic mail packages the data into coded form in
often offer the ability to encrypt messages. Specialized encryption devices can be such a way that only
authorized parties can access
No
inserted into the data-transmission media to encrypt all the data that passes
through. The level of encryption that you implement depends on the value of the the information.
c

data. When considering the value, consider what loss would be incurred if your
competitors or the general public were to become aware of the contents of the
data.
en
Do

Many vendors, both hardware and software, offer encryption devices. Data
encryption typically occurs at the Presentation layer of the OSI model.
Data can be encrypted using a variety of encryption methods. Three commonly
used methods are described below.
se
r

Think of encryption as
Encryption method Description coding a message using a
Stream Cipher Encrypts data a bit at a time. Each plain text bit is transformed into secret code.
fe

encrypted cipher text. These algorithms are relatively fast to

ea

execute. The cipher text is the same size as the original text. This
method produces fewer errors than other methods, and when errors
occur, they affect only one bit. cipher:
Block Cipher Encrypts data a block at a time, often in 64-bit blocks. It is usually A secret message or code.
Pl
Re

more secure than stream encryption, but is slower.

Padding When an uneven number of blocks is used to encrypt data, extra
ones, zeros, or both are added to the last block in the stream.

Lesson 14: Network Security and Fault Tolerance 349

 There are several modes of block cipher encryption. In ECB (Electronic
CodeBook) encryption, each block is encrypted by itself. Each occurrence of a
particular word is encrypted exactly the same. In CBC (Cipher Block Chaining)
encryption, before a block is encrypted, information from the preceding block is
added to the block. In this way, you can be sure that repeated data is encrypted
differently each time it is encountered. The CFB (Cipher FeedBack mode)
encryption model allows encryption of partial blocks rather than requiring full

ial
blocks for encryption. OFB (Output FeedBack mode) encryption is similar to
CFB except that shift registers are used in a different way.

Encryption Standards

er
Several standards have been developed for encryption. These include the Data
Encryption Standard (DES), RSA (Rivest, Shamir, and Adelman), and Digital
Signatures.

at
DES Encryption
DES encryption uses a symmetric key encryption method that employs a 56-bit

y
key on each 64-bit block. The U.S. government uses this encryption method and
has outlawed exporting it outside of the country. It was developed by IBM in the

op
late 1970s.
eM
tC
RSA Encryption
RSA encryption was developed at MIT by Ron Rivest, Adi Shamir, and Leonard
Adelman. This method uses a very large prime number as the public encryption
key. This key is then used to create a private key by processing the number
No
through a mathematical algorithm.
c

Digital Signatures
en
Do

A digital signature string can be added to messages to make sure that the message
wasn’t messed with. This method uses a public key to verify the message and a
private key to create the signature. A certificate authority computer contains a
″trusted application″ to validate the signature. While this method doesn’t neces-
sarily protect your information, you’ll know whether or not someone has
se
r

interfered with it.

fe
ea

Public Key and Private Key Encryption

In addition to scrambling the data, a key for unlocking the code is created. It
might be encrypted using either a public or private key encryption models.
Pl
Re

The private key or symmetric-key cryptography mode works with one key. All of
the objects on the network that have this key can encrypt and decrypt messages.
Because this key is only available to the sender and receiver of the message, it is
referred to as a private key. For security, the key must be kept safely guarded.

350
The administrator can establish the private key or it can be embedded in hard-
ware coding. If the key ever changes, all devices must be upgraded.
The public key or asymmetric-key cryptography mode is more secure than private
key because it uses two keys. The public key is available to everyone on the net- Public key encryption was
work, so messages are encrypted by using the recipient’s public key. Only the developed by Whitfield Diffie
recipient’s private key can be used to decrypt the message. This dual-key system and Martin Hellman;

ial
eliminates the need to share a private key. therefore public key
encryption is also known as
Diffie-Hellman encryption.

TASK 14C-1:
Discussing data encryption

er
1. What is data encryption?

at
2. Compare stream cipher and block cipher.

y
op
eM
3. List and describe some of the modes used in block cipher encryption.

tC
No
c
en
Do

4. How do DES, RSA, and Digital Signatures encrypt data?

se
r

5. Which encryption method uses symmetric key cryptography and which

uses asymmetric key cryptography?
fe
ea

6. What is another name for public key encryption?

Pl
Re

Lesson 14: Network Security and Fault Tolerance 351

 TOPIC 14D
Firewalls and Proxy Servers
Firewalls are specialized network devices used to limit access from outside net-

ial
works to your company’s network. These are used often in systems that connect
to the Internet. Such devices limit which users have access to both inside and
firewall: outside resources, and limit what functions those users can perform.
Specialized network devices
used to limit access from Proxy servers isolate internal network computers from the Internet. Users don’t
outside networks to your actually reach the Internet; the proxy server retrieves the requested Web page and
company’s networks. a copy of the page is passed to the user. The proxy server IP address is the only

er
address that goes outside of the network. Proxy servers can also be used to speed
up access to Web pages by retaining the requested page and serving that up to
subsequent user requests.
proxy server:
Server used to isolate

at
internal network computers
from the Internet. Firewalls
Firewalls control access between networks—both inbound traffic (entering your

y
network) and outbound traffic (leaving your network). Firewalls can be used to

op
prevent unauthorized access between intranets and the Internet or an extranet (an
eM
intranet external to your intranet). You can also configure firewalls to prevent
access to other intranets within your company (keeping marketing users out of the
R&D intranet).

tC
Controlling inbound access protects servers and resources on your intranet from
access by unauthorized Internet users. Controlling outbound access can be used to
limit your users’ access to Internet resources; for example, allowing access only
to work-related sites.
No
Firewalls contain packet filtering routers, circuit gateways, application gateways,
c

or a combination of these components. A combination is most effective because

each component operates at a different level of the OSI model, thus providing
access control at multiple levels.
en
Do

Packet filtering or screening routers restrict inbound traffic through source IP

address and port number filtering. These routers operate on the Data Link, Net-
work, and Transport layers of the OSI model. Depending on whether the filter is
configured to use port addresses, network addresses, or a combination of these,
will determine at exactly which layer or layers it operates.
se
r

Circuit gateways or circuit level proxies operate at the Session layer of the OSI
model. Outbound traffic is sent to a specific gateway port where software vali-
fe

dates a TCP or UDP session. After validating a session, a connection or gateway

ea

is opened. The gateway can substitute its own IP address for the requestor’s IP
address so the requestor’s address never leaves the intranet. Routing policies are
based on upper-level protocols such as HTTP.
The address translation feature of the IP gateways works at the Network layer of
Pl
Re

the OSI model.

The most secure firewalls are usually Application gateways or application-level
Border Manager from Novell proxies. These operate at the Application and Presentation layers of the OSI
is an example of a product model. Access control occurs at the Application level.
that can be both a firewall
and proxy server.

352
 ial
er
at
y
op
eM
Figure 14-4: Firewalls in relation to the OSI model.

TASK 14D-1:
Identifying the OSI layer where firewall components
tC
No
operate
c

Objective: Identify the layer of the OSI model where each of the compo-
nents described operate.
en
Do

1. Packet filtering or screening routers.

2. Circuit gateways or circuit level proxies.

se
r
fe

3. Address translation feature of IP gateways.

ea

4. Application gateways or application-level proxies.

Pl
Re

Lesson 14: Network Security and Fault Tolerance 353

 Proxy Cache Server
A proxy cache server or Web proxy cache server accelerates Web access perfor-
mance, reduces WAN traffic, and offloads Web server requests form the Web
server. This feature works at the Application and Presentation layers of the OSI
model.

ial
Web browsers and the source server (of Web server origin), which is accessed
over the Internet, usually communicate directly. Accelerating performance helps
make up for the delay or latency encountered in Web-based connections, and you
can use a proxy cache server to cache Web content to provide acceleration. This
enables browsers to retrieve data from a local source (the proxy server).
The proxy cache can be configured to use standard proxy cache, HTTP accelera-

er
tion, or hierarchical proxy cache.
Configuring the proxy server as a standard proxy cache server caches the most
often requested Web pages locally. This enables clients to retrieve pages locally
rather than each client needing to access the same page accessing it across the

at
Internet. This results in quicker response times for those users accessing the
cached copy of the page and in reduced WAN traffic.
You can also configure proxy servers as a public front-end to Web servers on

y
your private intranet. This feature is referred to as an HTTP accelerator or
reverse proxy. The Web server is replaced with a dedicated proxy cache on the

op
eM
proxy server. This eliminates running out of Web server connections, which can
cause bottlenecks and slow response times. All of the static data that is accessed
is cached by the accelerator so only dynamic data requests are actually served
from the Web server.

tC
Hierarchical proxy cache or network accelerators place multiple copies of the
cached content on second-level or peer servers. Requests not available on the
local proxy server are passed to the parent server. If the parent server can’t fulfill
No
the request, it is passed to peer servers (and if necessary, on out to the Internet).
After the data is retrieved, the proxy server stores it locally and passes the results
c

to the client. This can reduce access time by at least ten times in most cases.
en
Do

TASK 14D-2:
Discussing firewalls and proxy servers
Objective: Identify the term described in each statement.
se
r

1. Specialized network devices that are used to limit access from outside
networks to your company’s network.
fe
ea

2. Servers that isolate internal network computers from the Internet.

Pl
Re

3. The most secure firewalls that operate in the upper layers of the OSI
model.

354
4. Accelerates Web access performance, reduces WAN traffic and offloads
Web server requests from the Web server.

5. Places multiple copies of the cached content on second-level or peer

servers.

ial
APPLY YOUR KNOWLEDGE 14-1 Suggested time:
10 minutes

er
Reviewing security
In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.

1. Load the Skilled Assessment component of the Network+ Certification

at
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

y
3. Select all items for the Security topic, uncheck the Test Time check box, and

op
then click OK.
eM
4. Work through the sample questions and score the test.

tC
5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.
No
6. Exit the test.
c

TOPIC 14E
en
Do

System Fault Tolerance

System Fault Tolerance (SFT) increases the reliability of networks. The seven
SFT features we’ll examine are disk mirroring, disk duplexing, Hot Fix, read-
after-write verification, duplicate FATs and DETs, Transaction Tracking System,
se
r

and UPS monitoring.

fe
ea

Disk Mirroring
With disk mirroring , blocks of data written to one disk are simultaneously writ-
ten to another disk. Should one disk fail, the other can continue to operate
Pl
Re

without loss of data or downtime. Mirrored disks share the same disk channel.
Therefore, a failure in the channel would cause a failure in both mirrored disks.

Lesson 14: Network Security and Fault Tolerance 355

 ial
er
at
y
op
eM
Figure 14-5: Disk mirroring.

Disk Duplexing
tC
With disk duplexing , identical operations are performed on two disk drives
No
located on separate disk channels. Should one disk or disk channel fail, the other
c

disk can continue to operate without loss of data or downtime.

en
Do
se
r
fe
ea
Pl
Re

356
 ial
er
at
y
op
eM
Figure 14-6: Disk duplexing.

UPS Monitoring
tC
A uninterruptible power supply (UPS) is valuable in providing network reliability.
No
When power goes out, this device provides power to the devices connected to it. uninterruptible power
c

This helps reduce or eliminate server data loss due to power outages. It can also supply (UPS):
limit or prevent damage to your servers connected to the UPS during power A device that provides
surges or brownouts. Usually only servers and server-related devices such as rout- backup power when the
en

ers are connected to UPSs. Workstations are not routinely connected to UPSs; electrical power fails or
Do

therefore, users are likely to still lose data in workstation RAM during a power drops to an unacceptable
voltage level.
outage.
UPSs can be online or off-line models. With an online UPS, power always flows
through the UPS to the devices connected to it. It is always actively monitoring
power as it provides power to devices. This has the added benefit of the UPS
se
r

functioning as a line conditioner, reducing or eliminating surges and brownouts to

the attached equipment. With an off-line UPS, the power is monitored by the UPS
fe

and is activated only when there is a drop in the power, resulting in a slight delay
ea

before becoming active. Power is not usually lost because the UPSs quickly
activate.
UPS monitoring enables the NOS to monitor a UPS so that users can be alerted
Pl

to log out and the operating system can be shut down properly if there is a power
Re

outage.

Lesson 14: Network Security and Fault Tolerance 357

 Using UPSs with Windows NT
Windows NT 4.0 provides software that will communicate with many types of
UPSs. When Windows NT receives a signal that the power has gone out, it
schedules a proper shutdown. You can also configure the UPS monitoring pro-
gram to broadcast a message and run the programs you specify. Configure the
UPS monitoring software with Control Panel UPS utility. Check your UPS docu-

ial
mentation for specific settings.

er
at
y
op
eM
tC
No
Figure 14-7: UPS support.
c

Using UPSs with NetWare

en
Do

UPS monitoring enables NetWare to monitor a UPS so that NetWare can alert
users to log out and shut down the operating system properly if there is a power
outage. If your UPS connects to your server through a serial port, use
AIOCOMX.NLM. After loading AIOCOMX, load UPS_AIO. It is recommended
that these commands be added to your AUTOEXEC.NCF files so that they are
se
r

loaded every time the server is started. A UPS_AIO status screen informs you of
the UPS status and activity.
fe

When loading UPS_AIO, you can specify the following parameters:

ea

Parameter Description
DOWNTIME=n The length of time in seconds to wait before shutting down the
Pl
Re

server. If power comes back on before this time is up, the server isn’t
shut down. If the UPS battery runs down before this time is up, the
server is shut down at once. The default is 300 seconds.
MSGDELAY=n The length of time in seconds before the first UPS broadcast
message (including the time remaining until shut down) is sent out
to users. The default is 5 seconds.

358
Parameter Description
MSGINTERVAL=n The length of time in seconds between the automatically generated
UPS broadcast messages that are sent to users. The default is 30
seconds.
DRIVERTYPE=n THE AIO driver type can be 1, 2, or 3. The default is 1, which is
used by AIOCOMX. Refer to documentation for a UPS using a

ial
different driver for the driver type number.
BOARD=n Specifies the AIO board number as set by the driver manufacturer.
When using AIOCOMX, the number is displayed upon loading. If
you are using another driver, check the documentation for the board
number.
PORT=n Specifies the port number as set by the driver manufacturer. When

er
using AIOCOMX, the number is displayed upon loading. If you are
using another driver, check the documentation for the port number
SIGNAL_HIGH If your UPS uses high values to learn if power is off or if the battery
is low, use this parameter. Most systems use low values, in which
case you won’t need to use this parameter.

at
y
Using UPSs with Linux

op
eM
You can also connect a UPS to your Linux workstation or server. See the Hard-
ware Compatibility HOWTO to make sure the UPS you have chosen is
compatible with Linux.

tC
A network-aware daemon can be run to shut down and start up client or server
computers that are connected to UPSs.
No
Sector Sparing
c

Sector sparing is a system in which every time the operating system reads or
writes data to the disk, it checks the integrity of the sectors to which the data is
en

being written. If a problem is detected, the data is moved to another sector and
Do

the problem sector is marked as bad. Bad sectors won’t be reused.

Sector Sparing in Windows NT

se

Windows NT Server automatically uses sector sparing in systems that use SCSI
r

disks. Sector sparing isn’t available on ESDI or IDE disk drives.

fe
ea

NetWare’s Hot Fix

With Hot Fix , bad blocks from a disk’s main storage area are redirected to a
Pl

portion of the disk called the Hot Fix Redirection Area . Hot Fix supports SCSI,
Re

ESDI, and IDE drives. The bad blocks are marked so that they will not be used
again.

Lesson 14: Network Security and Fault Tolerance 359

 ial
er
at
y
op
eM
Figure 14-8: Sector sparing (NetWare’s Hot Fix).

NetWare Specific Fault-tolerance Features tC

No
While the above fault-tolerance features can be found on many different network
c

operating systems, the following features are unique to the NetWare environment.
en

Read-after-write Verification
Do

After a block of data is written to a hard disk, it is read back from the hard disk
and compared to the original data in memory. If, after several attempts, data read
from the hard disk does not match the data in memory, NetWare stores the data
in a block in the Hot Fix Redirection Area and marks the bad block so that it will
se

not be used again.

r
fe
ea
Pl
Re

360
Duplicate FATs and DETs

ial
er
at
y
op
eM
Figure 14-9: Duplicate FAT and DET files.
If a portion of a FAT or DET becomes damaged, files might be damaged or lost.tC
To reduce the possibility of losing data this way, the NetWare operating system
No
maintains duplicate copies of FATs and DETs. If one copy of a table is damaged,
c

data can be retrieved by using the remaining table. If the table were damaged
because of a bad block, the table would be remapped by the Hot Fix feature, and
the repaired table would be updated from the good copy.
en
Do

Transaction Tracking System

Transaction Tracking System (TTS) provides the ability to back out transactions
(such as changes in a database file) that have been interrupted by the failure of a
se

network component. Suppose that a client has $800 in a savings account, $200 in
r

a checking account, and needs to write a check for $1000. For the check to clear,
the client needs to transfer funds from the savings account to the checking
fe

account. The process involves deducting the $800 from the savings account and
ea

adding the funds to the checking account.

During the transfer process, if power were interrupted on the computer that was
performing the transfer, the transaction would be lost. The result would be $0 in
Pl
Re

savings, and $200 in checking. With TTS, the last transaction would be backed
out, and the result would be $200 in checking and $800 in savings.

Lesson 14: Network Security and Fault Tolerance 361

 TASK 14E-1:
Examining system fault-tolerance features
Objective: Identify the fault-tolerance feature described in each of the
following steps.

ial
1. Identical operations are performed on two disk drives, each on a sepa-
rate disk channel.

2. Enables NetWare to monitor an uninterruptible power supply so that

er
NetWare can alert users and shut down properly if there is a power
outage.

at
3. Blocks of data written to one disk are simultaneously written to another
disk on the same channel.

y
4. Backs out transactions that have been interrupted by the failure of a

op
eM
network component.

tC
5. Blocks from a disk’s main storage area are redirected to another portion
of the disk.
No
6. After a block of data is written to a hard disk, it is read back from the
c

hard disk and compared to the original data in memory.

en
Do

7. Prevents a damaged server table from making data inaccessible.

APPLY YOUR KNOWLEDGE 14-2

se

Suggested time:
r

20 minutes
Configuring disk mirroring
fe

In this activity, you’ll complete a simulation on the Network+ Certification

ea

CD-ROM that simulates configuring disk mirroring on a Windows NT system.

1. Load the Activities component of the Network+ Certification CD-ROM.

Pl
Re

2. Click on the Network Design, Implementation And Maintenance link; then,

click on the Fault Tolerance link.

3. Point to the Configure Disk Mirroring link. Read the scenario, and then click
on the link.

362
4. Complete the steps in the Activity as prompted in the lower-right corner of
the screen.

5. When you’re finish the Activity, exit the Network+ Certification CD-ROM.

RAID

ial
A Redundant Array of Inexpensive Disks (RAID) provides disk fault tolerance.
There are actually many types, or levels, of RAIDs. These data protection
schemes have been available for many years. The following table lists the types
of RAIDs and indicates if Windows NT Server or NetWare supports each particu- RAID:
lar level: Redundant Array of

er
Inexpensive Disks. An
arrangement of multiple disk
Included drives and a sophisticated
RAID Included with Windows with controller to offer higher
level Features NT Server NetWare performance or reliability, or
both, than a single disk

at
0 Striping Yes No
drive.
1 Mirroring and duplexing Yes Yes
2 Striping with error correction No No

y
3 Striping with parity on a single drive No No
4 Striping by block with parity on a No No

op
single drive
eM
5 Striping with parity information spread Yes No
across drives

Striping (RAID Level 0) tC

No
Striping is the process of spreading data across multiple drives. In RAID Level 0,
c

data is striped in small chunks across multiple drives. (Windows NT Server uses
64K blocks.) By spreading data across multiple drives, striping can dramatically
improve read and write performance. However, because the data is spread across
en

multiple drives, the risk of data loss is significantly increased. If any one of the
Do

drives fails, you will lose all of your data. Striping makes no provision for recov-
ering or rebuilding data. For this reason, striping is generally considered a
performance-enhancement feature rather than a fault-tolerance feature. Windows
NT Server provides built-in support for RAID Level 0. You must have at least
two physical disk drives, or up to a maximum of 32 disk drives, to implement
se
r

striping on a Windows NT Server system. NetWare doesn’t support this level.

fe
ea
Pl
Re

Lesson 14: Network Security and Fault Tolerance 363

 ial
er
at
y
op
eM
Figure 14-10: Striping with RAID Level 0.

Mirroring (RAID Level 1)

tC
Mirroring duplicates an entire partition on another partition. Duplexing is a varia-
No
tion of mirroring, and requires an extra disk drive controller card. When you use
c

either scheme, all of the data and drive information from one partition is dupli-
cated on the second partition. Any partition can be mirrored, including the
System and Boot partitions. To implement mirroring, you must have two physical
en

disk drives. To implement duplexing, each of the two physical disk drives must
Do

be connected to its own disk controller. Windows NT Server supports both mir-
roring and duplexing across SCSI, ESDI, and IDE disk drives. Mirroring is an
expensive fault-tolerance solution; it uses twice as much disk space as would be
used for data storage without mirroring. Duplexing is even more expensive; you
will also need a second disk controller. However, RAID Level 1 is the only fault-
se

tolerance solution you can use to protect your boot disk.

r
fe
ea
Pl
Re

364
 ial
er
at
y
op
eM
Figure 14-11: Mirroring and duplexing.

Striping with Error Correction Code (RAID Level 2) tC

Striping with error correction code spreads data, bit by bit, across multiple drives.
No
Error correction code information is built from the bits and stored on a separate
c

drive. This scheme isn’t often used. Neither NetWare nor Windows NT Server
offer this RAID level.
en
Do

Striping with Parity on a Single Drive (RAID Level 3)

In a RAID Level 3 system, data is spread byte by byte across multiple drives.
Parity information is stored on a separate drive. A RAID Level 3 system requires
at least three, and usually no more than five, drives. RAID Level 3 systems pro-
se
r

vide both performance and fault-tolerance enhancements.

One example of its performance enhancement is that, in a three drive system, the
fe

amount of data in a file to be saved is divided across two of the drives. Poten-
ea

tially, the file will be written (or read) twice as fast as in a single drive system. If
a drive fails in a RAID Level 3 system, the information that was on that drive
can be rebuilt from the remaining drives and the parity drive. In fact, many
Pl

hardware-based RAID Level 3 systems simply log the event of a drive failure and
Re

continue operating. Then, at your convenience, you replace the failed hard drive.
Some RAID Level 3 systems even allow you to install a new drive while the sys-
tem is up and running (sometimes called hot swapping). Neither NetWare nor
Windows NT Server offer this RAID level.

Lesson 14: Network Security and Fault Tolerance 365

 Striping by Block with Parity on a Single Drive (RAID
Level 4)
A RAID Level 4 system spreads data block by block across multiple drives. A
block refers to whatever the block size is on the disks. Usually, blocks are groups
of one to sixteen disk sectors. Parity information is stored on a separate drive. A

ial
RAID Level 4 system uses at least two, and usually no more than five, drives.
RAID Level 4 systems provide both read-performance and fault-tolerance
enhancements. Potentially, the system can read as many single-block-sized files as
there are drives at one given time. However, because a single parity drive is used
and every write must be accompanied by an associated parity write, only one file
at a time can be written. As with RAID Level 3, if a single drive in a RAID

er
Level 4 system fails, the data can be rebuilt from the remaining drives and the
parity drive. Neither NetWare nor Windows NT Server offer this RAID level.

Striping with Parity Spread Across Multiple Drives

at
(RAID Level 5)
RAID Level 5 systems operate in much the same manner as RAID Level 3

y
systems. However, instead of storing parity information on a single drive, the par-
ity data is spread across drives. Because both data and parity are spread across

op
eM
the drives, more than one read and write can occur simultaneously. Hardware-
based RAID Level 5 systems offer many of the same features as hardware-based
RAID Level 3 systems, such as continuous operation, the ability to schedule
failed-drive replacement, and even the ability to hot-swap failed drives. Windows

tC
NT Server provides built-in software support for this RAID level, calling it ″strip-
ing with parity.″ You need at least three, and can have as many as 32, physical
disk drives to implement striping with parity spread across multiple drives. The
system and boot partitions can’t be part of a stripe set with parity across multiple
No
drives. They must be stored on a separate partition. NetWare doesn’t offer this
c

RAID level.
en
Do
se
r
fe
ea
Pl
Re

366
 ial
er
at
y
op
Figure 14-12: Striping with parity.
eM
TASK 14E-2:
Discussing RAID
1. Which RAID level is also referred to as disk mirroring?
tC
No
c

2. Which RAID level is also referred to as disk striping without parity?

en
Do

3. Which RAID level should you use for the greatest level of fault toler-
ance? Why?
se
r
fe

4. Which RAID level would best suit your needs? Why?

ea

5. How many disks are required for RAID level 5?

Pl
Re

Lesson 14: Network Security and Fault Tolerance 367

 Windows NT File System Fault Tolerance
Windows NT uses disk partitions to identify storage areas. We’ll look at the use
of volume sets and stripe sets as file system fault-tolerance features for FAT and
NTFS file systems.

ial
Working with Disk Partitions
A partition is a logical area of disk space that you can format and treat as a
single storage unit. There are two types of partitions. Primary partitions are parti-
tions that you can use to boot the computer, whereas extended partitions are
partition: partitions that you can further subdivide into logical drives. Any one physical

er
A logical area of disk space disk can contain a maximum of four partitions. Of the four, one can be an
that you can format and treat extended partition. An area of a disk that isn’t included in a partition is called
as a single storage unit.
free space.
The Disk Administrator utility is the main tool you’ll use to examine and work

at
with disk partitions in Windows NT. You must be a member of the Administrators
group to run it.

y
Working with a Volume Set

op
eM
A volume set combines 2 to 32 areas of free space and 1 to 32 physical disks into
a single, larger logical drive. Space within the volume set is filled in contiguously
beginning with the first area of space on the first drive. Volume sets create more

tC
usable-sized volumes from small fragments of free space and also conserve drive
letters. Areas on different physical types of drives can be combined into one vol-
volume set: ume set; however, a volume set cannot contain either the boot or the system
The combination of 2 to 32 partition. If you boot to other operating systems, the volume sets will be acces-
No
areas of free space on 1 to sible only if that other system supports them.
32 physical disks into a
c

single, larger logical drive.

en
Do
se
r
fe
ea
Pl
Re

368
 ial
er
at
y
op
Figure 14-13: A volume set.
eM
To use Disk Administrator to create a volume set, [Ctrl]-click to select multiple
areas of free space; choose Partition→Create Volume Set; and format the new

tC
partition.
To delete a volume set, select any area in the set and choose Partition→Delete.
(You cannot delete just a portion of the set.)
No
If an existing primary partition, logical drive, or volume set is formatted to
c

NTFS, you can add additional free space to the drive to extend it. To extend the
drive, [Ctrl]-click to select the existing drive and the areas of free space you want
to add, and choose Partition→Extend Volume Set. The new area of the drive will
en

be formatted the next time you restart the computer.

Do

TASK 14E-3:
Discussing volume sets
se
r

1. What utility is used to create a volume set?

fe
ea

2. What is the minimum number of disks required to create a volume set?

Pl
Re

Lesson 14: Network Security and Fault Tolerance 369

 3. Can a volume set contain a boot or system partition?

4. Can you resize a volume set?

ial
Working with Stripe Sets
A stripe set combines 2 to 32 areas of free space and 2 to 32 physical disks into
one logical drive. Unlike a volume set, you must have at least two physical disks

er
to implement a stripe set. Also, in a stripe set, the areas of free space must be
about the same size. If not, then on each of the drives, Disk Administrator will
use areas of space that are approximately equal to the smallest available area.
stripe set: (For example, if you select one 100 MB section and two 400 MB sections of free
The combination of 2 to 32 space, Disk Administrator will create a stripe set of about 300 MB.) A stripe set
areas of free space on 2 to can improve disk performance by writing information evenly to each drive in the

at
32 physical disks into one set. It allows for concurrent input and output operations across all the disks in the
logical drive. set.

y
op
eM
tC
No
c
en
Do
se
r

Figure 14-14: A stripe set.

fe
ea

As with a volume set, disks in a stripe set can be of different physical types.
However, the set cannot include the boot or system partition, and it will not be
accessible if you boot to another operating system that does not support stripe
sets. As you would with a volume set, you must format the stripe set and restart
Pl
Re

the computer before accessing the space in the new set.

To implement a stripe set:
1. Start the Disk Administrator.
2. Select the areas of free space on each of the drives that you want to include
in the stripe set.

370
3. Choose Partition→Create Stripe Set.
4. Disk Administrator will display the range of valid sizes for the stripe set
based on the free-space areas you have chosen. Enter the size you want in
the Create Stripe Set dialog box.
5. Click OK.

ial
The entire set will be assigned to one drive letter.

TASK 14E-4:
Discussing stripe sets

er
1. Name one benefit stripe sets have over volume sets.

at
2. Name two restrictions on stripe sets that don’t apply to volume sets.

y
op
3.
eM
Describe a scenario where you might implement a stripe set.

APPLY YOUR KNOWLEDGE 14-3 tC Suggested time:

No
20 minutes
Configuring disk striping with parity
c

In this activity, you’ll complete a simulation on the Network+ Certification

CD-ROM included with this course.
en
Do

1. Load the Activities Component of the Network+ Certification CD-ROM.

2. Click on the Network Design, Implementation And Maintenance link.

3. Point to the Configure Disk Striping With Parity link. Read the scenario, and
then click on the link.
se
r

4. Complete the tasks in the Activity as prompted in the lower-right corner of

fe

the screen.
ea

5. When you have completed the Activity, exit the Network+ Certification
CD-ROM.
Pl
Re

Lesson 14: Network Security and Fault Tolerance 371

 TOPIC 14F
Backing Up Data
Consider the value of the data stored on your network. A week’s worth of

ial
changes and additions to files or to a database can have greater value to a com-
pany than the entire network on which it is stored.
As companies use networks for more and more of their business communications
and data storage, the value of the information kept on these networks increases
dramatically. The loss of any of this information can be devastating to a
company. Therefore, it is essential that companies back up this information.

er
You can use the backup utilities included with the operating system, or you can
purchase third party backup products. Many organizations have developed backup
software to address this need.
When you plan the implementation of a backup system, you should consider the

at
following questions:
• What backup media will you use?

y
• Does the system give you the performance and reliability you need?
• Is the system flexible and easy to use?

op
•
eM
In the event of a disaster, will the recovery be complete?
• How fast can the system be up and running again?

Backup Media
tC
There are more and more choices every year when it comes to backup media. A
No
few years ago you only had a choice between reel-to-reel tapes, QIC cartridges,
c

and very expensive DAT recorders. Today, the costs of the DAT recorders and
media are within the range of most networks. You can also choose to use optical
drives, recordable CD-ROMS, or additional disk drives. For workstation backups,
en

you might consider using Iomega’s Jaz or Zip disks.

Do

Magnetic tape is still the most popular backup media. These are most commonly
found as described in the following table.

Maximum storage
Media sizes Description
se
r

DAT (Digital Audio At least 1 GB, up to Used by many different size networks; 4mm tape,
Tape) 12 GB about the size of an audio tape
fe

DLT (Digital Linear At least 10 GB, up to Used mainly by mid to large size networks;1⁄2-inch
ea

Tape) 12 GB cartidges
QIC (Quarter inch At least 40 MB, up to Original width was 1⁄4 inch; available in 3 1⁄2-inch
Cartridge) 25 GB (Traven) or 5 1⁄4-inch cartridges; usually used by
Pl
Re

smaller networks

372
Procedural Backup Policy
When you plan your backup methods, it is important to establish a policy and to
set up procedures to be followed. Some considerations include:
• Hardware —choosing the appropriate hardware for your environment.
• Assignment of backup administration —who is responsible for performing

ial
the backup?
• Backup frequency —evaluating the cost of potential data losses and estab-
lishing an acceptable minimum backup frequency.
• Backup types —choosing the appropriate backup method for your
organization.

er
• Backup set —the number of tapes (or other media) required to complete a
backup.
• When the backup is performed —evaluating the best time to run the backup,
ensuring that users are logged out, and ensuring that the greatest percentage
of files are not in use.

at
• Data identification —labeling backup media and storing it in the safest pos-
sible location.

y
• Data testing —occasional testing of backup data.
•

op
Maintenance schedule —hardware and media maintenance (or replacement).
•
eM
Recovery operation plan —laying out a specific plan for the complete recov-
ery of lost data.
• Backup methods —the structure of how backup media are rotated in and out

tC
of the backup schedule.

Rotation Method and Backup Types

No
c

Every system administrator’s schedule varies; however, companies generally

maintain a minimum daily backup to update new and changed data. Different cri-
teria help to determine the best strategy for backing up data.
en
Do
se
r
fe
ea
Pl
Re

Lesson 14: Network Security and Fault Tolerance 373

 ial
er
at
y
op
eM
tC
Figure 14-15: The grandfather backup method.

Rotation Methods
No
c

A common, secure backup rotation method is the grandfather method. One

backup set is designated for Mondays, one for Tuesdays, one for Wednesdays,
and one for Thursdays. These backup sets are reused on the same day the follow-
en

ing week. Every Friday during the month, you use a new backup set. The next
Do

month, these Friday backup sets are reused in the same order. At the end of each
month, a new backup set is used. Depending on your needs, these monthly
backup sets can be reused the following year or kept as a permanent record and
replaced with new backup sets.
Label each of the daily backup sets with the name of the day of the week; label
se
r

the weekly tapes with number 1 through 4 for each Friday; and label the monthly
tapes with the name of the month.
fe
ea
Pl
Re

374
 ial
er
at
Figure 14-16: The Tower of Hanoi backup method.

y
Another backup rotation method in use at some companies is the Tower of Hanoi
method. In this method, five media sets are used. Media set A is used every other

op
eM
day (2 days apart). Media set B is used every fourth day. Media set C is used
every eighth day. Media set D is used every sixteenth day. Media set E is alter-
nated with media set D. This doubles the backup history with each media set
used (2, 4, 8, or 16 days until the media set is overwritten). This enables you to

tC
have media sets with most recent versions of files (those media sets used most
frequently, such as sets A and B).
Label each of the media sets with a letter or number (media set 1 or A). You can
apply this rotation method to a daily or weekly rotation schedule. Five media sets
No
is the minimum required when performing a daily rotation, and eight media sets
c

are required for a weekly rotation.

You should archive media sets as needed. For example, you might save the E set
en

each month for a permanent archive. The tape sets you pull for archive will be
Do

based on your company needs.

NetWare Backup Types

se

The following table describes the different backup types supported by the
r

NetWare backup utility.

fe

Backup type Description

ea

Full All information is backed up, regardless of whether it has been

backed up before. The archive bit [A] is cleared.
Incremental New files and files that were created or modified since the last full
Pl
Re

or incremental backup are backed up. The archive bit [A] is cleared.
Differential All files that were created or modified since the last full backup are
backed up. The archive bit [A] is not cleared.

Lesson 14: Network Security and Fault Tolerance 375

 ial
er
at
y
Figure 14-17: Relative time to back up data.

op
eM
Although an incremental backup combined with a regularly scheduled full backup
is usually the fastest type of backup to perform, any time that you need to per-
form a full restore, you must restore the last full backup plus every incremental

tC
backup since the last full backup (in the same order they were backed up). This
process can be become very time-consuming.
In the long run, it might be more advantageous to perform differential backups,
combined with a regularly scheduled full backup, because all new or modified
No
files since the last full backup are included, and you would need only two media
c

sets to perform a full restore. Each differential backup following the last full
backup will take an increasing amount of time. Because differential backups are
based on the last backup that cleared the archive bit, you should not mix incre-
en

mental and differential backups. Differential backups count on the archive bit not
Do

being cleared. Any differential backups following an incremental backup would

not back up all of the modified files since the last full backup because the archive
bit was cleared with the incremental backup.
If you choose to perform full backups every time you back up your data, be
aware that a full backup takes the longest amount of time. On the other hand, it
se
r

is also the method that allows the fastest full restores, because you will need only
one media set to restore data.
fe
ea

Backup Types and Their Backup and Restore Time

Requirements
Pl
Re

The following table summarizes each of the three backup schemes and describes
how long a full backup and full restore will take for each scheme, measured in
relative time.

376
 Relative time necessary Relative time necessary to
Backup type to perform a full backup perform a full restore
Full backup only Longest Shortest
Incremental backup with Shortest Longest
full backup

ial
Differential backup with Time increases each day; Shorter than an incremental backup, but
full backup shorter than full backup longer than full backup.
overall, but longer than an
incremental backup.

Using Figure 14-16, the following table compares the number of tapes required to

er
restore data. The necessary tapes and amount of time for restoring data depends
on what is being restored and the type of back up that was used.

When restoring Full Incremental Differential

Full week 1 tape from Day 5 All 5 tapes Tapes 1 and 5

at
All data up to Day 3 1 tape from Day 3 Tapes 1, 2, and 3 Tapes 1 and 3

y
To perform a full backup, the

op
Storage
eM User object that performs the
backup must have Read and
It is advisable to keep backup media in a safe location, preferably in a locked, File Scan rights to the file
fire-safe room. In addition, you should consider moving at least one full backup system and the Browse

tC
per week to an offsite location. Many companies offer this service and store your object and Read property
tapes in a locked and fire-safe area. Consider how critical your data is when you rights to NDS. The user must
decide how often to move backups offsite. A disadvantage of offsite storage is also know any passwords
that, when you need to restore data, you will have to involve a third party in assigned to file servers or
workstations. To restore the
delivering the media back to your location. This process can be time-consuming.
No
file system or NDS tree, the
c

user must have Create rights.

NT Backup Methods
en

The following table lists NT backup methods and their descriptions. A good
Do

backup system for an NT system will probably combine several of these methods.
You’ll also need to consider factors such as the amount of tape needed for a
given backup, the time available for the backup, the time available for the restore,
and the degree to which current backups are necessary for your network.
se
r

Windows NT
Backup method backup Description
fe

Full backup Normal Backs up selected files and marks them as

ea

backed up. This method includes backing up

files that have not changed since the last
backup.
Pl

Copy Copy Backs up selected files without marking them

Re

as backed up.
Incremental backup Incremental Backs up only selected files that have changed
since the last backup, and marks them as
backed up.

Lesson 14: Network Security and Fault Tolerance 377

 Windows NT
Backup method backup Description
Daily copy Daily copy Backs up any selected files that have been
modified that day, without marking them as
backed up.
Differential backup Differential Backs up only selected files that have changed

ial
since the last backup, without marking them as
backed up.

TASK 14F-1:

er
Discussing backup strategies
1. How many tape sets are required when using the grandfather rotation

at
method?

y
op
2.
eM
How many tape sets are required when using the Tower of Hanoi rota-
tion method?

3. tC
List and describe the NetWare backup types.
No
c
en
Do

4. List and describe the NT backup methods.

se
r
fe
ea
Pl
Re

378
APPLY YOUR KNOWLEDGE 14-4 Suggested time:
1 hour
Reviewing Basic Knowledge
In this activity, you’ll work through some sample test questions that deal with the
information presented in the course so far.

ial
1. Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

3. Select all items for the Basic Knowledge topic, uncheck the Test Time check

er
box, and then click OK.

4. Work through the sample questions and score the test.

5. If you want to review the questions and correct answers, click Review Items.

at
Use the Explanation and Next buttons to work through the answers; then use
the Scores button to return to your test score window.

y
6. Exit the test.

op
eM
Summary
In this lesson, you examined a wide range of network security and fault-
tolerance features. First, you looked at security models. NT has user and
share level security whereas NetWare and UNIX have only user level
tC
No
security. In learning about password security, you learned about what consti-
c

tutes a secure password. Data encryption was examined and you discussed
some of the techniques used to encrypt information. Next, you examined the
use of firewalls and proxy servers in protecting your network. In Topic E,
you examined some of the system fault-tolerance features including disk
en

mirroring and duplexing, ups monitoring, sector spanning, and NetWare spe-
Do

cific features such as read-after-write verification, duplicate FATs and DETs,

and TTS. You also looked at the various RAID levels that have been
defined. Windows NT supports its own file system fault-tolerance features
which includes volume sets and stripe sets. Finally, you looked at backup
methods and procedures for NetWare and NT networks.
se
r
fe
ea
Pl
Re

Lesson 14: Network Security and Fault Tolerance 379

 LESSON 14 REVIEW
14A What are the levels of security in a NetWare network?

ial
What are the levels of security in a Windows environment?

14B Describe the features of a good password.

er
14CWhy might you want to encrypt data?

at
y
op
14D Compare firewalls and proxy servers.
eM
tC
14E List some of the system fault-tolerance features available on NT and
NetWare networks.
No
c

Which RAID levels are included with NT Server and which are included
with NetWare?
en
Do

14F List some of the backup media choices available today.

se
r
fe
ea
Pl
Re

380
Preparing for Network LESSON
Installation
15

ial
Data Files:
Network+ Certification
Overview CD-ROM

er
This lesson describes the things you need to know and do before you begin Lesson Time:
installing a new or updated network operating system. 2 hours, 30 minutes

at
Objectives
To make sure that you have the information necessary to complete a successful

y
network installation, you will:

op
15A
eM
Review basic pre-installation requirements.
This topic identifies some of the information you’ll need to gather from
other sources before installing network hardware or a network operating

tC
system.

15B Determine the effect of environmental factors on computers and

networks.
No
You will identify some of the actions that need to be taken before install-
c

ing network hardware or a network operating system.

15C Describe physical network components.

en

This topic covers some common equipment found in many computer

Do

networks.

15D Identify potential compatibility problems that can affect the success
of your network installation.
This topic describes some connection, upgrade, and cabling pitfalls you
se
r

should know about so that you can make sure they don’t wreak havoc on
your network.
fe
ea
Pl
Re

Lesson 15: Preparing for Network Installation 381

 TOPIC 15A
Planning and Gathering Information
One of the most important facets of any large project is the attention given to

ial
planning and gathering information. Whether you are an in-house network admin-
istrator or an independent network consultant, preparing an implementation
strategy for any company that has decided to install a new network or upgrade an
existing network can be pretty overwhelming. Using a detailed course of action
and correctly assessing the needs and expectations of the company can help make
the task a little less daunting.

er
Pre-installation Information
Before you install a network operating system, there are some things that you

at
need to know about. Getting this information ahead of time helps you save time
during the actual installation, and can prevent problems from arising after the
installation is completed. At a minimum, you should get the following:

y
• An administrative account and password
•

op
eM
At least one test account and password
• Network addressing information
• Company SOPs that apply to network use

Administrative Accounts and Passwords

tC
For most network operating systems, you need to use an administrative account
No
and password to upgrade NOSs, create user accounts, set up print services, and
c

install network applications and additional hardware.

The NetWare admin account,
the Windows NT
en

Administrator account, and

Do

the UNIX root (superuser)

account all provide full
access to the NOS by
default; however, the
administrative account
assigned to you might have
se

a different name.
r
fe
ea
Pl
Re

382
User Accounts and Passwords
Because the administrative account usually provides access to the entire network,
you should also use ″regular″ user accounts to test network access. Test accounts In any situation where you
might be actual user accounts, or they can be accounts created specifically for need to use an administrative
testing purposes. Try to test as many different configurations as you can to ensure or user account and
that all users have the correct access to the network resources (file storage, appli- password, make sure that

ial
cations, printers, Internet access, and the like) that they need. you keep the passwords to
yourself. And don’t forget to
let the administrator know
when you are finished with
Network Addressing Information the accounts, so that the
passwords can be reset.
Gathering information about network addresses before you start installing the net- These actions help maintain

er
work is crucial, especially if you’ll be using TCP/IP. If the company already has the integrity of the system,
and they help reduce your
a network, get a list of all IP addresses—all addresses assigned to the company, liability in case of future
all addresses assigned to existing network devices, as well as all address that security breaches.
haven’t yet been assigned. You’ll also need to know the existing subnet masks
and how the addresses are assigned (manually or by DHCP).

at
Company SOPs

y
Most companies have Standard Operating Procedures (SOPs) that must be

op
eM
followed. To be sure that your installation follows existing corporate standards,
review the company’s SOPs for policies and procedures that are relevant to the
computing environment, such as: SOP:
Standard Operating

tC
• Naming standards for network components, including users and groups. Procedure. SOPs outline the
• Accessibility standards for users, applications, and physical resources, basic administrative
including storage allocation and access to network files, printers, and workings of a company, from
hiring guidelines to rules
applications. controlling the use of
No
• Internet access guidelines and restrictions, including hours of use and company resources during
c

restricted or forbidden sites. non-work hours.

• Email use guidelines, including restrictions on personal use.

en

• User administration, including the distribution of administrative duties and

Do

associated rights among various IT personnel.

TASK 15A-1:
se
r

Identifying information required for network

installation
fe
ea

1. Explain the types of network accounts you need for a network installa-
tion and the uses for each type.
Pl
Re

Lesson 15: Preparing for Network Installation 383

 2. Why should you gather information about network addressing before
you install a network component?

ial
3. What are some network issues that might be covered by a company’s
SOPs?

er
4. You are asked to create several new user accounts on an existing net-
work prior to performing a network upgrade. Which of the following
best describes the minimum information you need to be able to perform
this task, and why?

at
a. Copies of company SOPs and a list of available network addresses.
b. A valid IP address and subnet mask.

y
c. Copies of company SOPs and an administrator account.

op
eM
d. The users’ full names and department numbers, and the network
resources they need to access.
e. An administrator account and a list of valid IP addresses.

tC
5. You are a consultant working with a manufacturing company that wants
to upgrade the NOS running on the corporate network. The company’s
liaison gives you a project notebook containing a temporary security
badge, a comprehensive list of employee names and phone numbers, the
No
name and password for an administrative account, a detailed floor map,
c

documentation for the production process, and the naming standards

documents for the current version of the NOS. Will you be able to per-
form the upgrade successfully? Why or why not?
en
Do
se
r
fe
ea
Pl
Re

384
TOPIC 15B
Environmental Issues
Before you begin installing any network hardware, inspect the site and identify

ial
where servers, hubs, switches, routers, printers, and workstations will be placed.
The effectiveness of computer equipment can be limited by any of the following
environmental factors:
• Room conditions, such as heat, humidity, and dust.
• The placement of personal items, such as portable heaters and other portable

er
electronic devices.
• The distance between the computer equipment and the electrical and other
outlets needed to run the equipment.

at
Physical Conditions
Heat and humidity have a huge impact on computer performance, especially

y
servers. Dust and other small airborne particles can also affect computers.
• When a room is too warm or too cold, chances are that the computers in that

op
room simply won’t function correctly.
eM
• Extreme humidity levels, whether high or low, can damage computer and
other electronic components. High levels of humidity can cause moisture to

tC
form on computer parts, creating short circuits that can permanently damage
them. Low humidity increases the static electricity in a room, which can also
ruin computer components.
• Dust, cigarette smoke, and other particulate matter can get inside computers
No
and clog connections, causing intermittent problems for the user.
c

Temperature Precautions
en
Do

A good rule of thumb is that if people will be uncomfortable in a room, the com-
puters in the room will probably be ″uncomfortable,″ too. Maintaining a
temperature of 70° Fahrenheit will prevent computer components from
overheating. Many companies maintain a climate-controlled server room to help
with this issue. If the server room’s air conditioning is kept separate from the rest
se
r

of the building, and is powered by a dedicated electrical outlet, there will be

fewer instances of trouble, especially during heat spells when central air can fail.
Also, make sure that you allow computer equipment to reach the ambient tem-
fe

perature before you turn it on. For example, if you need to move a device from
ea

one location to another, you should let the device warm up or cool down to
match its new location. Moving equipment from cool to warm surroundings can
Pl
Re

Lesson 15: Preparing for Network Installation 385

 also cause a condition known as chip creep. The expansion and contraction
caused by the temperature changes actually causes the chips to loosen in their
sockets.
chip creep:
A condition caused by You can install warning monitors that will alert you (by audible alarm, beeper,
repeated moving of computer broadcast message, or other means) if the temperature in the server room is out-
equipment to different side of the acceptable temperature range that you set.

ial
temperatures. Changes in
temperature cause the
components to expand and
contract, which can in turn Electrostatic Discharge Precautions
cause the connections
between a chip and its Whenever two objects are joined and then separated, static charges are generated.
socket to become loose. The separation of the objects causes the transference of electrons from one object

er
to another, making one of the objects negatively charged and the other positively
charged. When either of these objects come near to or in contact with another
object that doesn’t have the same charge, discharge occurs to balance the
disparity.
Electrostatic discharge (ESD) can be devastating to electronic components. Elec-

at
tronic components can be damaged by discharges of as little as 20 or 30 volts
(For you to feel the snap of a static discharge, about 3,000 volts of static electric-
ity have to cross from your body to a ground or potential source.) Only the most

y
sensitive instruments can detect low levels of static charges. These charges can
cause equipment failure or degradation at any time, even during production of the

op
eM
components. A component that’s degraded by ESD might pass inspection and fail
later on, usually when the component is installed into a computer. This makes it
easy to underestimate the potential for damage that ESD can cause.

tC
Although ESD problems are common, they’re avoidable by following some basic
precautions. The benefits of following an ESD prevention plan include:
• Less downtime
• Fewer spare parts needed and used
No
•
c

Fewer intermittent problems to track down

• Fewer service calls
• Fewer unhappy customers
en
Do

To prevent damage from static discharge, use the following guidelines:

• Use an ESD kit. ESD kits usually contain wrist straps and mats. The wrist
strap electrically connects your body to the unit you’re working on, such as
a computer. The wrist strap is designed to slowly transfer any static charge
between you and the device. After a few seconds of wearing the strap, you
se
r

can safely work with the components. Periodically check that the ground
cords are secured to the mat and to the computer.
fe

Warning: You must unplug the device from the electrical socket before
ea

Don’t use a wrist strap when

attaching the wrist strap to prevent getting an electrical shock from a faulty
you are working on component or wiring system. To prevent static discharge, you need only be
monitors! The high voltage concerned with grounding yourself to the component, not to the entire elec-
of monitors can travel
Pl

trical system!
Re

through the wrist strap to

you. • Don’t touch components or integrated chips by the electrical leads.
• When you are working on boards that contain integrated circuits, make sure
that others do not touch you.

386
• Use static-shielding bags to store and move boards and integrated circuits.
Make sure that the bags don’t have any holes in them.
• Keep nonconductors, such as plastic, Styrofoam, and synthetic and silk
clothing, away from computers and components. Static-shielding bags are
• Monitor the humidity in areas that will hold computers and components. different than anti-static
ESD problems occur more often in low-humidity environments. Maintaining bags. Static-shielding bags

ial
the humidity at 40% to 60% will help alleviate ESD problems. are usually grayish silver,
while anti-static bags are
• Don’t place components on conductive surfaces, such as metal or static bags. usually pink or blue. Anti-
static bags do not provide
• Check all areas for proper ESD practices. For example, you should check the same level of protection
that field personnel wear proper clothing (no synthetics or silks, and no as static-shielding bags,
metal jewelry), that the service site has static-discharge equipment available, because they don’t protect

er
that shipping, receiving, and storage areas are kept at the proper humidity. their contents from external
static sources.
• Avoid having carpeting, or if this is impossible, make sure that anti-static
mats are available.

at
Cleanliness Precautions
As you check for environmental issues, don’t forget that dust, dirt, and the like

y
can get inside computer equipment and cause many problems. Cigarette smoke
and dust particles are large enough to damage high-precision devices such as hard

op
eM
drives. A deposited layer of these particles can also act as an insulator, trapping
heat inside chips or computers. Whenever possible, place computers in areas that
aren’t subject to a lot of dust and other airborne matter. Filtering air sources is
Recommended humidity
another possible solution for this problem. In circumstances where this isn’t pos-

tC
levels vary. Some sources
sible, such as on a manufacturing floor, you should make arrangements for cite relative humidity levels
periodic cleaning of the components. as low as 30% and as high
as 60% and above. The best
bet is to check the vendor’s
No
recommendations.
Personal Items
c

In many companies, employees can personalize their working space. In situations

like this, check areas that will house workstations, and verify that there aren’t any
en

electrical devices that can affect the performance of the workstation. Some of the
Do

most notorious of these items include space heaters, radios, and portable televi-
sion sets. Although space heaters can help keep employees comfortable during the
colder months, they can cause problems with computer equipment. In addition to
the temperature fluctuations that can occur in areas with space heaters, you also
need to watch out for power fluctuations caused by the heaters as they cycle on
se
r

and off during the day, as well as the electromagnetic interference that can be
caused by the heater’s motor. If a workstation is plugged into the same outlet as a
space heater, data can be corrupted on the hard drive or on the network cable.
fe

Radios, portable televisions, and cellular phones can also cause power fluctua-
ea

tions, but, more importantly, they are sources of radio frequency interference,
which can corrupt data on the network cable. Finally, check for miscellaneous
office equipment, such as paper shredders, calculators, and postage machines, that
Pl
Re

can cause similar problems for workstations and network communications.

Lesson 15: Preparing for Network Installation 387

 Spatial Conditions
Check floor plans to see how close the network equipment will be placed in rela-
tion to the available electrical outlets. Avoid excessive use of extension cords and
power strips when possible, as these can contribute to power fluctuation
problems. Although many power strips also claim to have surge protection capa-
bilities, the fact is that most surge protectors are set too high to really prevent

ial
damage to your computers. Try to keep all computer equipment on outlets that
are not used by other devices, and install uninterruptible power supplies (UPSs)
where necessary.

TASK 15B-1:

er
Reviewing environmental issues
1. List some environmental conditions you should check before installing

at
network hardware.

y
2. Explain ESD and how you can prevent it from affecting computer

op
eM
equipment.

tC
No
c

3. Why should you keep radios and TVs away from computer equipment?
en
Do

4. You are reviewing a site for an upgrade installation. This company is

quite casual, allowing employees to dress and to decorate their cubicles
any way they want, and even allowing them to bring their pets to work
with them. What suggestions would you make to help improve the suc-
cess of the upgrade?
se
r
fe
ea
Pl
Re

388
5. Which of the following locations would be your choice for placing a file
server, and why?
a. In an enclosed wiring closet next to the main telephone switch
panel.
b. In a ventilated office with a southwest view.

ial
c. In a corner of the plant floor.
d. In an air-conditioned room that contains other electronic devices.

TOPIC 15C

er
Networking Components and Connectors
To be effective and efficient while you are installing network hardware, it’s

at
imperative that you be able to recognize network components and their
connectors. This section describes some of the most common network compo-
nents and connectors.

y
op
Network Components
eM
Network Interface Cards (NICs)

tC
Network interface cards (NICs) enable you to connect computers together to form
a network. NICs resemble other internal PC cards and are usually ISA- or PCI-
based. NICs are also specific to a network topology (such as Ethernet or Token-
Ring) and often a connection type, such as BNC, RJ-45, and IBM data network interface card
No
connectors. A Token-Ring media filter can be used to connect a Token-Ring NIC (NIC):
A printed circuit board that
c

to a UTP-based Token-Ring network.

plugs into both the clients
(personal computers or
workstations) and servers,
en

Patch Panels connecting them to the

Do

transmission medium, and

Patch panels are central wiring points for devices on a UTP network. Patch pan- controls the exchange of data
els don’t contain any electronic components; they’re used only to organize the between them. Also referred
to as a network interface
workstation cables and provide a way to connect those cables to switches, routers, board or network adapter.
and hubs. Make sure that you meet or exceed the rating for existing cabling—
se

using a Category 3 patch panel with Category 5 wiring decreases the network’s
r

rating to Category 3. On UTP Ethernet networks, patch panels have RJ-45

connections. patch panel:
fe

A group of sockets that acts

ea

as a manual switching center

between incoming and
outgoing lines in a
communications, electronic,
Pl
Re

or electrical system.

Lesson 15: Preparing for Network Installation 389

 Patch Cables
Patch cables connect a network device to the main cable. Make sure that you
include the length of all patch cables in any calculations regarding segment
length. On UTP networks, patch cables have RJ-45 connectors.
patch cable:
A short length of cable

ial
(usually no longer than 10
feet) used to connect ports in Repeaters
patch panels or in expansion
boards and systems that are Repeaters amplify the signal on a cable, enabling you to extend the maximum
in close proximity. Also distance of a network segment. Repeaters have two or more connections, one for
referred to as a patch cord. each network segment. They operate only in the Physical layer, so although you
can use them to connect different media types, they don’t convert protocols.

er
repeater:
A communications device Hubs
that amplifies or regenerates
the data signal to extend the Hubs connect workstations in a star topology. Generally used in 10BASE-T and

at
transmission distance. 100BASE-T networks, hubs are usually one of three types: passive, active, and
intelligent. Passive hubs simply provide a physical connection so that all stations
can communicate, so they don’t require a power source. Active hubs are powered

y
devices that repeat signals and are also known as multiport repeaters .Intelligent
hub: hubs are those active hubs that also contain configuration and management

op
A central connecting device
eM
options. Most hubs have several RJ-45 connections.
in a network that connects
communications lines in a
star configuration.

tC
Bridges
Bridges connect network segments of similar type (Ethernet to Ethernet, or
multiport repeater: Token-Ring to Token-Ring) or dissimilar type and that forward only those data
A hub in a 10BASE-T packets that are not addressed to a device on the local segment. Bridges operate
No
network is often known as a at the Data Link layer of the OSI model. Like repeaters, bridges have two or
c

multiport repeater, because it more connections, but they don’t have to be of the same type.
sends any input signal to all
outputs. Also referred to as
en

an active hub.
Routers
Do

Routers interconnect networks by reading the source and destination address of a

bridge: packet and forwarding the packet based on the information it has gathered about
A device that connects two the networks it’s connected to. Routers can also be used to connect dissimilar
LAN segments and forwards networks (such as Ethernet and Token-Ring) so that data can be transferred from
se
r

data to the segment that it’s one to another. Routers can contain one or more connections of dissimilar types.
addressed to.
fe
ea

router:
A device that forwards data
packets from one local area
Pl
Re

network (LAN) or wide area

network (WAN) to another.

390
Brouters
Brouters combine the functions of a bridge and a router and are used in cases
where both routable and nonroutable protocols exist on a network. Routable pro-
tocols (such as IP) are routed, while nonroutable protocols (such as NetBEUI) are
bridged. Brouters can have two or more connections of dissimilar types. brouter:
A device that combines the

ial
function of a bridge and a
router. Also referred to as a
Print Servers bridging router.

Print servers control the management of network print services. Print servers can
be hardware, software, or a combination of the two. Hardware print servers can
be dedicated machines (file servers that run only print services), separate entities

er
such as the HP JetDirect print server or Intel NetPort print server, or they can be print servers:
integrated into other network components, such as the HP JetDirect NICs. Print A computer or hardware
server connections include not only the connections necessary to connect to the device that controls network
print services.
network such as BNC or RJ-45, but also connections for the printers that the
print server will manage. Printer connections can be 25-pin parallel, 9-pin serial,

at
or 25-pin serial.

y
UPSs

op
eM
UPSs provide continuous, reliable power to connected computers. UPSs contain
high-end batteries and power inverters. When plugged into a wall outlet, the UPS
uses the line current only to charge the batteries in the unit. The connected com-
puter gets its power only from the batteries. In the case of a power outage, the

tC
battery continues supplying power to the equipment. When the outage is
extended, some UPSs can even provide for a graceful shutdown of the connected
equipment. UPSs generally only provide electrical connections. Some UPSs also
contain line conditioners to help eliminate power sags, spikes, and surges.
No
c

Peripherals
en

Peripherals are devices that aren’t part of a computer but are directly attached
Do

(via a serial, parallel, or USB peripheral port) to the computer, such as printers,
scanners, external modems, external speakers, backup devices, and removable
storage devices such as CD-ROM drives or bays and Zip or Jaz drives. Peripher- peripheral:
als can be connected to either servers or workstations, and can often be shared as Any hardware device
network devices. Some peripherals can even be connected directly to the network connected to a computer,
se

such as a monitor, keyboard,

r

cable. Peripheral connections can be of many types; for example, many external printer, disk, tape, graphics
CD-ROM drives are designed to use the SCSI (Small Computer System Inter- tablet, scanner, joy stick,
face) interface, but there are several types of SCSI connectors available, such as
fe

paddle, and mouse.

DB-25 female, Centronics-50, and the high-density 50-pin SCSI-II and 68-pin
ea

SCSI-III connectors.
Pl
Re

Connectors
There are many different types of connectors and connections for computer
equipment. Some of the most common connectors and their uses are described in
this section.

Lesson 15: Preparing for Network Installation 391

 BNC Connectors
BNC connectors are most commonly used with coaxial cable on 10BASE 2
networks. They’re round, with a half-turn locking shell mechanism that prevents
the cable from disconnecting.

ial
D-type Connectors
D-type connectors are also known as D-sub or DB connectors. They’re the oldest
type of network connector, and are used to connect many peripherals to PCs (via
the serial and parallel ports). D-type connectors are so called because they are
shaped like the letter D turned on its side. These connectors can be either male

er
(containing rows of pins) or female (containing rows of sockets), and contain
small screws at each end to prevent the cable from disconnecting. The original
implementation of D-type connectors in a network environment was the Attach-
ment Unit Interface (AUI) connector used with DIX Ethernet—these connectors
are still included with some Ethernet NICs for compatibility. You can use an AUI

at
and external transceiver to switch media types, such as coax to UTP.

y
Fiber-optic Connectors

op
eM
There are several types of fiber-optic connectors available, including D-4,
ESCON, MIC, SC, ST, and SMA. The following table describes these connectors
and their uses.

tC
Connector Description
D-4 Uses a threaded nut connector for attachment with ceramic ferrules to hold the
fiber, and can be used for single-mode or multimode cable.
No
MIC Stands for Medium Interface Connector. MIC connectors are usually found in
c

FDDI networks. They are single, keyed connectors that hold two fibers to
ensure that the male and female ends of the connection will connect properly.
ESCON Stands for Enterprise System CONnection (Architecture). ESCON connectors
en

are commonly used in FDDI networks, and are similar to MIC connectors,
Do

except that they have a retractable cover that makes it easier to attach it to a
transceiver.
SC Stands for Subscriber Connector. SC connectors are one-piece, push-on
connectors with two receptacles for sending and receiving. Using two fibers
ensures that the correct fiber is plugged into the appropriate jack. SC
connectors work well for splicing fiber-optic cables because they make a very
se
r

strong connection. Cables must be polished and perfectly aligned for use with
this connector.
fe

ST Stands for Straight Tip. ST connectors have a ceramic jacket glued to the fiber
ea

core, with the fiber sticking out of the end of the jacket. The outer shell
resembles a BNC connector and locks onto the jack with a quarter twist. The
end of the fiber must be polished and perfectly perpendicular to its length.
SMA Stands for Sub-Miniature Assembly. SMA connectors are similar to ST
Pl
Re

connectors, using a threaded outer shell to stabilize connections in vibration-

prone environments. The 905 version uses a straight ferrule design, and the
906 uses a stepped ferrule with a plastic sleeve for a more precise alignment.
Each fiber strand uses one connector.

392
IBM Data Connector
The IBM data connector is a genderless attachment mechanism where two con-
nectors click together to establish a connection between a Token-Ring lobe and a
MAU or patch panel. To maintain the ring structure inside the MAU, the IBM
data connector is self-shorting, meaning there’s a circuit across it even when IBM data connector:
nothing’s plugged in. A device used to connect

ial
IBM Token Ring stations by
using Type 1 STP cable.

Peripheral Connectors
Most peripherals are connected to computers via the following connectors:
•

er
Serial connectors are usually 9-pin or 25-pin D-type (DB-9 and DB-25).
• Parallel connectors are usually 25-pin D-type (DB-25), but some printer
manufacturers have adopted a 36-pin Centronics connection, so most printer
cables have a DB-25 connector at one end and a Centronics connector at the
other end.

at
• PCMCIA slots accept 68-pin connectors.
• SCSI-I connectors are usually a 50-pin connector similar to a Centronics

y
printer connection or a 25-pin D-type (DB-25) connector. Newer SCSI
implementations use high-density connectors (50-pin for SCSI-II and 68-pin

op
for SCSI-III).
eM
• USB connectors are either rectangular (Type A) or square (Type B).

TASK 15C-1:
Identifying network components
tC
No
c

1. Compare and contrast repeaters and hubs.

en
Do
se
r

2. List some network components that use DB-25 connectors.

fe
ea

3. In what order would the following components usually be connected?

Pl

Explain your response.

Re

a. Workstation, NIC, patch cable, wall jack, network segment, patch

panel, hub, network backbone, hub, patch cable, patch panel, patch
cable, file server.

Lesson 15: Preparing for Network Installation 393

 b. Workstation, UPS, patch cable, wall jack, network segment, patch
panel, patch cable, router, network backbone, router, patch cable,
patch panel, patch cable, file server.
c. Workstation, NIC, patch cable, wall jack, network segment, patch
panel, patch cable, hub, network segment, bridge, network segment,
patch panel, patch cable, file server.

ial
d. Workstation, NIC, patch cable, hub, patch cable, wall jack, network
segment, patch panel, network backbone, hub, patch cable, file
server.

4. Which of the following sets of connectors can be used for SCSI devices?
Explain your response.

er
a. BNC, DB-25, and Centronics-50.
b. Centronics-36 and Centronics-50.
c. DB-9 and Centronics-50.

at
d. DB-25 and Centronics-50.

y
APPLY YOUR KNOWLEDGE 15-1

op
Suggested time:
30 minutes
eM
Visually identifying network components
Visually identify the following network components. If you’d like, you can sketch

tC
each component in the space provided. Use all available resources, including this
course material, Web sites, the Network+ Certification CD-ROM , and equipment
found in the classroom, and note where you found the component.
No
1. BNC connector.
c
en
Do

2. Bridge.
se
r

3. Brouter
fe
ea
Pl

4. Centronics connectors.
Re

394
5. DB-9 connector.

6. DB-25 connector.

ial
7. Hub.

er
8. IBM data connector.

at
9. Network interface card.

y
op
10. Patch panel and patch cable.
eM
11. Print server.
tC
No
c

12. Repeater.
en
Do

13. RJ-11 connector.

se
r
fe

14. RJ-45 connector.

ea
Pl
Re

15. Router.

Lesson 15: Preparing for Network Installation 395

 16. SC connector.

17. SCSI connectors.

ial
18. ST connector.

er
19. UPS.

at
20. USB connectors.

y
op
eM
TOPIC 15D
Compatibility Issues tC
No
As you inspect a site for a potential network installation or upgrade, you also
c

need to be aware of the following issues:

• Cabling and connection compatibility.
• Hardware and software compatibility.
en
Do

• User expectations.

Cabling and Connections

se
r

Check the site for existing cabling, and carefully document what you find. If a
local modem will be used, make sure that an analog telephone line is available
and plainly marked as such. Plugging an analog modem into a digital phone jack
fe

Remember that many phone can easily destroy the modem.

ea

systems and network cabling

both use RJ-type connectors, Also make sure that you know what type of cable is attached to a wall data jack.
with phone systems using RJ-45 jacks can be used with both Ethernet and Token-Ring cabling.
RJ-11 and network cabling
Pl
Re

using RJ-45. Because these Finally, remember that patch cables contribute to the overall length of a network
are so similar, it’s actually segment. If the difference between the workstation to the wiring closet is in ques-
quite easy to mistakenly plug tion, use a cable tester to verify the overall cable length.
an RJ-11 modem cable into
an RJ-45 digital phone or
data jack.

396
Hardware and Software
Most network installations aren’t built from scratch. You’ll usually have to deal
with existing equipment. Verify that the new hardware is compatible with the
existing hardware, that existing software will run on the new hardware, and that
all networking components are compatible.

ial
It’s best to do a complete hardware and software inventory prior to any
installation. For the software inventory, include the name, version number, and
serial or license number, along with vendor support information and any appli-
cable Web sites.

er
User Expectations
You can prevent a lot of service calls and complaints by verifying that the users
of the network understand the nature of the installation or upgrade you’re doing.
Often, when users hear that they are getting a new network or new hardware,

at
they expect—and rightly so—that all their programs will work perfectly fine
under the new system. The only way to verify this is to do a complete inventory
of all user applications and test them on the new system.

y
By checking with users, you can also learn about issues that affect the existing
network. Users might expect that all existing problems will be solved by the

op
eM
installation you’re planning, but if you don’t know about the problem, it’s highly
unlikely that the new installation will solve it. Document and deal with all exist-
ing problems before you complete the installation; then verify that the new
system doesn’t recreate the old problems or cause new ones.

TASK 15D-1: tC
No
c

Discussing compatibility issues

1. Before you connect a new modem to the telephone system, which of the
en

following should you do? Explain your response.

Do

a. Make sure that the wall jack has two connections so that a phone
can also be connected.
b. Make sure that the modem is placed within arm’s length of the
wall jack.
se
r

c. Make sure that the modem is compatible with the workstation.

fe

d. Make sure that the wall jack is connected to an analog phone line.
ea

2. You’re running UTP cable in an Ethernet environment from a worksta-

tion to a hub, and then to a wall jack. Which of the following are
included in the overall cable length? Explain your response.
Pl
Re

a. Main cable run.

b. Hub patch cable.
c. Workstation patch cable.

Lesson 15: Preparing for Network Installation 397

 d. Hub internal wiring.

3. You are a consultant who has been asked to replace a workstation in the
Customer Service Department. When you arrive in the department with
the new computer, the user begins a litany of complaints about the old
computer. What should you do before replacing the workstation?

ial
er
Suggested time:
APPLY YOUR KNOWLEDGE 15-2

at
30 minutes
Preparing for network installation
In this activity, you’ll work through some sample test questions that deal with the

y
information presented in this lesson.

op
1.
eM
Load the Skill Assessment component of the Network+ Certification
CD-ROM. (Remember to accept the disclaimer.)

tC
2. Start the Network+ test.

3. Select all items for the Implementing the Installation of the Network topic,
and uncheck the Test Time check box, and then click OK.
No
4. Work through the sample questions and score the test.
c

5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
en

and answers; then use the Scores button to return to your test score window.
Do

6. Exit the test.

se
r

Summary
fe

In this lesson, you’ve determined what types of information are necessary

ea

before beginning a network installation, reviewed various networking com-

ponents and connectors, and identified environmental and other issues that
can affect the success of a network installation or upgrade.
Pl
Re

398
LESSON 15 REVIEW
15A List some information you should gather before installing a new or
upgraded network operating system.

ial
15B Describe a good environment for computers.

er
15CWhat is an advantage of using UPSs with file servers?

at
y
15D What types of compatibility issues should you review before performing
an installation or upgrade?

op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 15: Preparing for Network Installation 399

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Maintaining the Network LESSON
16

ial
Overview Data Files:
This lesson provides you with the information you’ll need to keep a network Network+ Certification
up and running smoothly. CD-ROM

er
Lesson Time:
3 hours
Objectives

at
To be able to monitor and maintain a network, you will:

16A Identify various sources of network support documentation.

y
You’ll discover some places where you can get information about
network-maintenance issues.

op
eM
16B Review network backup concepts and strategies.
You’ll look at the information that makes up the network backup compo-

tC
nent of a disaster recovery plan.

16C Investigate how to protect a network from viruses.

You’ll find out about the types of viruses and virus protection software.
No
16D Discuss the importance of updating network software.
c

In this topic, you’ll see how patches, fixes, and other software updates
can help prevent problems on a network, and you’ll review the process of
en

downloading and applying a patch file.

Do
se
r
fe
ea
Pl
Re

Lesson 16: Maintaining the Network 401

 TOPIC 16A
Network Documentation
Many forms of documentation exist to help you maintain computers and

ial
networks. Because of the growing complexity of computer systems, it’s vital that
you know where to find information that relates to the hardware and software that
you need to support. Most major hardware and software vendors provide several
types of documentation to assist in installing and maintaining their products,
including paper documentation and online documentation, either via a CD-ROM
or over the Internet. Telephone support is also available for most products, but

er
usually entails paying a fee for its use. Although much of this information comes
directly from the vendors, there are other sources, too. Many publishing houses
specialize in producing technical reference books for a wide variety of software
packages.

at
Vendor Documentation

y
Vendor documentation comes in many forms. When you buy a product, some
form of documentation (paper, CD-ROM, or Web pages) is bound to be included,

op
eM
but this might not be the most current information available. When you’re prepar-
ing to implement a new product, you should check the following sources:
• FAQ lists. If you have never used a type of product before and want to get
some background information, FAQ lists can be helpful. They’re usually

tC
developed by the vendor and contain many of the installation and basic sup-
FAQ: port issues you’ll need to be aware of when you’re working with the
Frequently Asked Questions. product.
•
No
Hardware compatibility lists. Before you recommend or install a network
component, check the vendor’s hardware compatibility list to be sure the
c

new component will work with existing equipment and software. For
example, neither Windows NT Server nor some versions of NetWare can be
installed on just any computer, so consulting the hardware compatibility list
en
Do

can prevent installation nightmares. Be aware, though, that the various items
on the list might not have been tested for compatibility in combination with
other products, so you might still need to verify with the vendor that the
products will work together. For Linux, check the How To hardware compat-
ibility list for your components. If components aren’t listed, they might or
se

might not work, but you can always write your own drivers if none are
r

available.
• Installation and migration guidelines and tools. When you’re installing a
fe

new NOS on an existing network, performing the actual installation might

ea

end up being the easiest part of the job. Once the new NOS is up, you need
to get the old data onto the new system. This can be particularly hard when
you’re switching from one NOS to another, or you’re installing a major revi-
Pl
Re

sion of the same NOS. Most vendors provide migration tools to assist
network personnel in completing this task successfully. At a minimum, the
vendor might have a migration document that can guide you through the
process.

402
• White papers. White papers exist for many technical subjects. Check for
white papers dealing with installation or implementation, particularly those
that deal with real-world scenarios—they can give you insight on what to
expect during your implementation of the product. white paper:
A detailed technical report
• Release notes and readme files . Release notes can tell you what’s changed written by a vendor,
in a software revision, while readme files often outline known issues with consultant, research entity, or

ial
the shipping product, along with other changes that occurred after the origi- other organization that
nal (paper) documentation was finalized. Review both of these documents to describes a specific topic of
interest.
make sure you’re installing the most recent version of the product, and to
make sure that you have the most up-to-date information available.

release notes:

er
Finding the Documentation A document that describes
the history of a product, and
There are many places to find vendor documentation. Check the following items identifies the changes
to make sure you’ve got the most current information and the information that implemented in specific
product versions.
applies to your specific situation:

at
• Product package. Although it’s probably not the most current or most spe-
cific information available, you should still review the docs that came with
the product. Most vendors provide user guides, installation manuals, and

y
other docs as part of the original purchase of the product. Increasingly, many
vendors are providing these docs in an electronic format instead of paper

op
eM
manuals. The manuals are often supplied on a CD-ROM, with only selected
documentation being reproduced on paper. Take the time to review the
manuals, and pay special attention to the minimum hardware and software

tC
requirements, before you start installing the product. The supplied docs
should also contain the information you’ll need about contacting the vendor
for additional support, including Web sites, email addresses, and telephone
support, as well as what information you’ll need to provide during a support
No
instance, such as version and serial numbers.
•
c

Internet. One of the quickest and most comprehensive sources of information

is the Internet. Virtually all vendors have some sort of online presence, and
there are many third-party sources of information available, too, from Web
en

sites such as www.supportsource.com and www.zdnet.com, to online user

Do

groups and newsgroups such as Iguana (Independent Group of Unix-Alikes

& Networking Activists) and comp.sys.novell (a Usenet group dealing with
NetWare). Because of the wealth of information on the Internet, it’s often
hard to find exactly what you’re looking for. Take some time to browse a
vendor’s Web site—usually they’ll have a Documentation or Support link on
se
r

the home page—and review available help regarding the search engine used
on the site. You can also try the more general Internet search engines such as
Infoseek or Google; however, you might need to use advanced searching to
fe

narrow down the information you receive.

ea

• Subscription services. Some vendors offer a paid subscription service that

includes periodically updated databases of support information deployed on
CD-ROM or DVD. Examples of these are Novell Support Connection and
Pl
Re

Microsoft TechNet. These programs often provide just about everything

you’d want to know about a product in one place, including FAQs, white
papers, utilities, support and troubleshooting issues, and software updates
and patches. Annual subscriptions usually cost less than $500 per year.

Lesson 16: Maintaining the Network 403

 • Third-party reference books. Unless you’re installing very new, proprietary
hardware or software on a network, it’s a safe bet that you’ll find at least a
few third-party books dealing with the product. For hardware, you’re prob-
ably not going to find a book devoted entirely to the piece of hardware you
bought, but there could be one that deals with a group of similar types of
products. For network operating systems, you’ll find literally hundreds of

ial
books dealing with NetWare, Windows NT, and UNIX. Larger bookstores
often have sections devoted to technical books, and you can also find many
of these books at the various online bookstores such as www.amazon.com
and www.barnesandnoble.com. Specialized book clubs, such as the Library
of Computer Information Services (www.lcis.com) also carry dozens of
books on networking hardware and software.

er
TASK 16A-1:
Gathering network documentation

at
Objective: To investigate some sources of network documentation.

y
Setup: You’ll need an Internet connection and a Web browser to com-
plete this task.

op
eM
1. Where’s the first place you should look for vendor documentation?

2.

tC
What types of information might you find on a subscription CD?
No
c

3. Find the following Web sites and record their URLs. If you want, briefly
describe the kinds of documentation available for future reference.

Welcome to Microsoft Product Support Services

en
Do

Microsoft TechNet Support

se
r

Microsoft TechNet CD Online Site - Home

fe
ea

Novell Customer Services

Pl
Re

Novell: Product Documentation

Novell Support Connection

404
 The Novell Support Connection CD

Novell: NetWare Users International

ial
Cisco Technical Assistance Center

CNET Help.com

er
SupportSource

ZDNet: Help & How-To

at
y
4. Discuss your findings with the rest of the class. Be sure to include your
thoughts on ease of use, the types of information you found, and when

op
you’d be likely to use these sites.
eM
TOPIC 16B
Backup Strategies tC
No
Although it’s impossible to be completely prepared for every natural disaster or
c

other occurrence that can crash your network, you should have a good backup
plan to minimize the impact of lost data. A backup plan is a set of guidelines and
schedules that outline the types of data that should be backed up and the fre-
en

quency at which they should be backed up. Good backup plans include answers
Do

to the following questions:

• What should be backed up?
• Where should the backups go?
• When should backups occur?
se

•
r

How often should backups occur?

• Who’s responsible for initiating and maintaining backups?
fe

• Where should the media be stored?

ea

• How often should backups be tested?

• What should be done when data is lost?
Pl
Re

Lesson 16: Maintaining the Network 405

 What to Back Up
The actual data that needs to be backed up regularly will vary according to a
company’s needs, but your backup plan should at least include measures for
backing up any data that is considered to be mission-critical. You also need to
consider whether backing up the server is enough, or if data on individual work-
stations also needs to be backed up. If users consistently save their important data

ial
files to the network, workstation backup is probably not as necessary as when
users save the majority of their data files to their local hard disks. Another thing
to consider is if you need to back up workstation configuration information.

Utilities

er
A backup utility is a software program that archives data on a hard disk to a
removable medium. Some utilities compress data before storing it, which can be
more efficient than simply copying data to the backup medium. Network operat-
ing systems usually contain built-in backup utilities, such as Microsoft’s MS

at
Backup and NT Backup, Novell’s SBACKUP, and the UNIX tar utility); how-
ever, you’ll probably find that these utilities don’t provide the same options as
most third-party backup utilities such as Computer Associates’s ARCServe and

y
Veritas’s (formerly Seagate) BackupExec.

op
eM
Media
The backup medium is the place where you store data that has been backed up.

tC
Among the most popular backup media are small- and large-capacity removable
disks, removable optical disks, and magnetic tape.
No
You can find background
Small-capacity Removable Disks
c

information about various

storage media at www.
pctechguide.com/storage. Small-capacity removable disks are magnetic disks that hold less than 500 MB
and that can be removed from disk drives and replaced as they get full. Although
en

they don’t hold a lot of data, they’re popular for partial workstation backup
Do

because of their low cost and ease of use, plus many newer computers come
equipped with one or more of these drives.

Large-capacity Removable Disks

se
r

Large-capacity removable disks store more than 500 MB of data; otherwise,

fe

they’re the same as small-capacity removable disks. Although they cost more than
ea

the small-capacity disks, they can hold more data and can easily be used for more
complete workstation backups. However, their use as server backups is limited
because most servers can’t be backed up on just one disk.
Pl
Re

406
Removable Optical Disks
Removable optical disks use a laser or other light source to read and write infor-
mation stored on disk. Examples of removable optical disks include CD-ROM,
CD-R, CD-RW, and DVD. They tend to have larger capacities (650 MB to 17
GB) than the magnetic removable disks, but they also tend to be slower. Optical CD-R and CD-RW are
jukeboxes, which can contain hundreds of disks, can expand capacities into the writable variations of the

ial
CD-ROM technology, while
hundreds of terabytes. There is also a hybrid category called magneto-optical
DVD is the next generation
(MO) disks that combine the benefits of both magnetic and optical storage of compact disc technology.
methods.

Tape

er
Tape backups are probably the most common, because tape drives are reliable,
inexpensive, and widely supported. Common tape backup formats include QIC,
Travan, DAT, DLT, and 8mm.
• QIC, or quarter-inch cartridge. The QIC format is one of the oldest formats

at
used. Its name is derived from the original width of the tape used (one-
fourth inch), although some implementations use wider and longer tapes.
There are two form factors for this type of tape backup—3.5-inch mini car-

y
tridges and 5.25-inch data cartridges. Capacities range from 100 MB to 25

op
GB.
eM
• Travan. The Travan format evolved from the QIC format, and provides
higher capacities by using wider tape, different tape guides, and improved
magnetic media. This technology is best used for small peer-to-peer

tC
workgroup backups. Capacities vary depending on the type of drive and tape
used, but generally range from 400 MB to 10 GB.
• DAT, or Digital Audio Tape. As the name implies, this tape was originally
designed as the next generation of audio tape, providing CD-quality sound in
No
a cassette tape format. DAT tape uses 4mm cartridges that conform to the
c

Digital Data Storage (DDS) standard and helical scan recording, which is
similar to how videotapes are recorded. DAT tapes are popular because they
provide high reliability and speed at a relatively low cost. Without compres-
en

sion, typical DAT cassettes can hold from 2 to 12 GB of data.

Do

• DLT, or Digital Linear Tape. DLT is a magnetic tape technology that uses
one-half-inch single-hub cartridges. It’s an adaptation of reel-to-reel record-
ing where the tape cartridge performs as one reel and the tape drive as the
other. DLT drives use the widest tape available and record data in a serpen-
se

tine pattern on parallel tracks grouped into pairs. Current drives record either
r

128 or 208 tracks.

•
fe

8mm. 8mm tape was first developed for use in the video industry. It’s similar
ea

to DAT tape in that it uses helical scan recording, but it tends to provide
greater capacities. Implementations of 8mm tape include standard 8mm,
Mammoth and Advanced Intelligent Tape (AIT), which provide capacities of
3.5 to 50 GB.
Pl
Re

Summary
The following table summarizes the backup media discussed in this section.

Lesson 16: Maintaining the Network 407

 Medium Category Capacity
Floppy diskettes Small-capacity removable 1.44 and 2.88 MB
SyQuest cartridges Small-capacity removable 44, 88, 200, and 230 MB
Iomega Zip disks Small-capacity removable 100 and 250 MB
Imation SuperDisk Small-capacity removable 120 MB

ial
Removable hard disks Small-capacity removable Variable
SyQuest cartridges Large-capacity removable 1 and 1.5 GB
Iomega Jaz disks Large-capacity removable 1 and 2 GB
Removable hard disks Large-capacity removable Variable
CD-ROM, CD-R, and Removable optical 650 MB to 1 GB
CD-RW

er
MO disks Removable optical 650 MB, 1.3 GB, 4.6 GB
DVD Removable optical 4.7 to 17 GB
QIC Magnetic tape 100 MB to 25 GB
Travan Magnetic tape 400 MB, 800 MB, 1.6 GB, 2.5 GB, 4 GB,

at
and 10 GB
DAT Magnetic tape 2, 4, and 12 GB
DLT Magnetic tape 35 to 70 GB

y
Standard 8mm Magnetic tape 3.5 to 14 GB

op
Mammoth
AIT
eM Magnetic tape
Magnetic tape
20 and 40 GB
25 and 50 GB

Schedules
tC
Setting up and following the proper backup schedule is just as important as
No
selecting the proper backup utility and medium. Although backup policies and
c

rotation schedules were discussed in detail in Lesson 14, we’ll review them again
here.
The three main types of backup are full, differential, and incremental:
en

•
Do

Full backups do just what the name says—they back up all network data,
regardless of the type of file or whether or not it has changed.
• Differential backups back up all network data that has changed since the last
full backup.
• Incremental backups back up all network data that has changed since the last
se
r

full or partial backup.

You should perform backups at a time when no users are logged in and no files
fe

are open. In most cases, this will be at night.

ea

Rotate backup tapes to keep costs down while maintaining the number of back-
ups necessary to restore critical data. Tape rotation strategies include daily (3-tape
method), weekly, the Tower of Hanoi method, and the grandfather method.
Pl
Re

408
• Daily. For a daily rotation, you can use three tape sets. On the first day of
the work week, perform a full backup on the first tape set. For succeeding
days in the work week, use the second tape set to back up modified files,
appending each day’s changes at the end of the tape set. At the end of the
work week, perform another full backup, store the first tape set offsite, and
erase all data on the second tape set. This method is best used when less

ial
than 25% of data changes on a daily basis.
• Weekly. For a weekly rotation, you label tape sets with the days of the week
and use a different backup tape set for each day. For example, if your com-
pany operates 7 days a week, you would label 7 tape sets and use the
Sunday tape set on Sunday, the Monday tape set on Monday, and so forth.
When the next Sunday arrives, reuse the Sunday tape set. This method is

er
simple to follow; however, you can’t restore data from more than one busi-
ness week in the past. Another weekly method uses six tape sets and is
similar to the daily method described above, with the main difference being
that you would use a separate tape set for each workday.
• Tower of Hanoi method. The Tower of Hanoi backup method uses 5 sets of

at
media, with the first being used every other day, the second used every
fourth day, the third used every eighth day, and the fourth and fifth used
every sixteenth day (alternately). This scheme can be used with daily or

y
weekly backups, but if you use it for weekly backups, you’ll need 3 more

op
tape sets.
eM
• Grandfather method. The grandfather method uses one tape set for each of
the first four days of the work week, one tape set for each of the end-of-
week backups, and one tape set for the end of the month.

tC
In addition, you should always make at least one full backup each week, clearly
label all tapes, and note each day that they’re used. Never exceed the maximum
usage guidelines provided by the tape manufacturer. And consider backing up
your backups by retaining at least a full backup each month.
No
c

Media Storage
en

You can store backup tapes onsite or offsite. Balance the need for quick recovery
Do

of data with disaster recovery needs. If you do store tapes onsite, make sure you
keep them in a locked, fire-safe room.

Testing
se
r

The most diligent backup strategy is useless if you can’t restore data from the
fe

backup tapes. Most backup utilities include a data-verification feature that you
ea

can use to ensure that the copy on tape matches the copy on the network. In
addition, you should also periodically restore data from a tape set (to a non-
production computer) to ensure that the data is readable.
Pl
Re

Lesson 16: Maintaining the Network 409

 Data Recovery
You can implement standard rules for users to follow when they need to request
data recovery. For example, you might want to assign one administrative user or
a group of users the responsibility of dealing with data restoration, particularly
those cases caused by users inadvertently deleting files that shouldn’t have been
deleted. Or you might implement a special form for users to complete when they

ial
request data recovery so that you can track the frequency of these types of
requests.

TASK 16B-1:

er
Discussing backup issues
1. What are some benefits of having a good backup plan?

at
2.

y
Which backup type requires no more than two backup sessions to fully
recover data? Explain your answer.

op
eM
a. Full.
b. Partial.

tC
c. Differential.
d. Incremental.

3. Which of the following correctly lists backup media in order of dimin-

No
ishing capacity? Explain your response.
c

a. Magnetic tape, removable optical disks, large-capacity

nonremovable disks, small-capacity removable disks.
en

b. Magnetic tape, large-capacity removable disks, removable optical

Do

disks, small-capacity removable disks.

c. Small-capacity removable disks, large-capacity removable disks,
removable optical disks, magnetic tape.
d. Magnetic tape, removable optical disks, large-capacity removable
se
r

disks, small-capacity removable disks.

fe
ea
Pl
Re

410
4. At least monthly, backup tapes should be:
a. Scanned for viruses.
b. Replaced.
c. Retained.

ial
d. Taken offsite.
e. Created.

5. What recording method is used to record on DAT tape?

er
6. You get a call on Tuesday from a user who can’t access the budget
spreadsheet she saved to the network. She apologizes for not calling
sooner, but she just returned from a long weekend. How can you help?

at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 16: Maintaining the Network 411

 TOPIC 16C
Virus Protection
Viruses are programs that run without your knowledge or permission and are

ial
designed to replicate themselves. Viruses normally inflict damage to your data or
programs, display messages, or slow system performance.
A Trojan horse is a program, similar to a virus, whose actual purpose differs from
its stated purpose. For example, a Trojan horse might mimic the features and
functions of a login command. In doing so, it would gather valid login names and
passwords that could be used later to compromise your data. Another type of Tro-

er
jan horse might initiate a denial of service attack by bombarding a network with
useless traffic. A Trojan horse doesn’t replicate itself.
Estimates put the number of viruses in the tens of thousands, with at least six
new viruses per day being released.

at
Types of Viruses

y
Viruses usually fall into one of two categories—boot-sector infection and file

op
infection.
eM
• Boot-sector viruses replace the code that normally resides in the master boot
sector of a disk with its own viral code. They’re spread by bootable floppy

tC
diskettes. When a system is booted with an infected disk, the virus loads into
virus:
A program that runs without memory, and all subsequent disks used on that system will be infected by
your knowledge or the virus. These viruses are no longer the primary source of infection. Symp-
permission, often causing toms of a boot-sector virus include receiving the message Missing Operating
damage to your computer or System or the message Hard Disk Not Found. Examples of boot-sector
No
data, and that replicate viruses include Stealth_C and Monkey B.
c

themselves so that they can

be spread to other • File infectors attach to executable files so that when the program is run, the
computers. virus is also run and loaded into memory. The virus then infects any other
en

programs that are run on the system. Examples of file infectors include
Do

Junkie and Die Hard 2.

Trojan horse: — Currently, macro viruses are the most common. Macro viruses are
A program that has a encoded into a macro that’s embedded into a document and are spread
different purpose than by sharing infected documents. The main access point for macro viruses
expected or stated. is as an email attachment through mail gateways and firewalls, although
se
r

they can also be spread by sharing infected files (via removable media
or shared network storage space). Symptoms include not being able to
fe

open or save files. Examples of macro viruses include Concept and

ea

Melissa.
— Worm viruses are also becoming rampant. Worms are viruses that can
replicate themselves and use memory, but don’t attach themselves to
other programs. Virus developers are having a field day creating worms
Pl
Re

based on groupware packages, such as Microsoft Outlook, that have

features such as data replication, encrypted databases, dynamic messag-
ing, and workflow agents. Threats also exist from Java applets and
Active-X objects. Examples of worms include W32/ExploreZip.worm
and W32/Pretty.Worm.

412
Virus Hoaxes
An offshoot of the threats caused by viruses is the proliferation of virus hoax
emails, such as Good Times, WOBBLER, and Join The Crew. Not only can for- There are several Web sites
warding these emails clog a mail server and network with unnecessary traffic, but that track virus and other
they also tend to undermine the vigilance of many users for when real virus alerts hoaxes; for example http://
are communicated. In addition, some developers have taken advantage of known www.snopes.com/

ial
spoons/faxlore/virus.htm
hoaxes to spread new viruses.
and http://www.kumite.
com/myths/myths/ .

Prevention
Education is often the key to preventing virus attacks on your systems. Inform

er
users about viruses and how they spread. Some companies limit access to bulletin
boards, restrict the use of non-approved programs (including games and
shareware), or use other methods to limit their exposure to the risk of viruses.
Regularly check your systems for viruses. Many commercial and shareware virus-

at
checking programs are available, and most computer periodicals review the
available products. Check the reviews and pick an anti-virus program that meets
your needs.

y
Most anti-virus programs check for viruses in one of two ways: First, anti-virus
programs can check the files on your system for virus signatures (program code

op
eM
fragments known to be part of a virus). Second, anti-virus programs catalog the
length of all your executable files. Later, the program can compare its stored
value with the current file sizes. virus signature:
The binary pattern of the

tC
Viruses are becoming more sophisticated. Some viruses are self-modifying, mak- machine code of a particular
ing them undetectable by programs that search for virus signatures. Other viruses virus. Anti-virus engines
actually seek out and disable anti-virus programs, such as Tremor, which detects compare their database of
when an anti-virus scan is in process and moves itself around in memory to virus signatures with the files
on the hard disk and
No
escape detection. As the viruses become more sophisticated, so do the detection
removable media (including
c
programs. the boot sectors of the disks)
Administrators should look out for hackers, scan their servers, and beware of as well as within RAM.
commercial Internet service groups, along with the other anti-virus practices they
en

observe.
Do

Protection
Although viruses are both widespread and dangerous, there are things you can do
se
r

to protect your network from infestation. Developing a virus-protection plan helps

to ensure that all parts of the network are checked for viruses and that the effects
fe

of virus infections are minimized. To implement an effective virus-protection

plan:
ea

1. Select an anti-virus program. There are hundreds of anti-virus software pack-

ages available, and most of them perform the same basic functions—
scanning disks, folders, and files to check for the presence of known viruses,
Pl
Re

notifying the user of the suspected presence of a virus, and eradicating any
viruses found. Most anti-virus programs are made up of two distinct

Lesson 16: Maintaining the Network 413

 components—the engine and virus definition (or signature) files. The anti-
virus engine is an executable file that performs the scanning, cleaning, and
notification processes. It uses the virus definition files to identify code seg-
virus definition file: ments that have been identified as being part of known viruses.
A file that lists known
viruses and how to clean 2. Identify where and when the anti-virus program should be run. For the most
them from computers. comprehensive protection, you should run anti-virus software on all com-

ial
Generally updated on a puter equipment—servers, workstations, standalone computers, and firewalls.
monthly basis. Also referred Even those machines that never connect to your network should be protected
to as virus signature file.
against viruses that can be spread by floppy disk and over the Internet or
other dial-up connections. You also need to consider the frequency at which
computers need to be scanned for viruses. In most situations, monthly scan-
anti-virus engine: ning is considered to be adequate, with additional scans performed in special

er
An executable file that uses circumstances, such as when you install the anti-virus software or if a virus
virus definition or virus outbreak is suspected.
signature files to identify the
presence of viruses on a 3. Determine how to keep virus signature files up-to-date. Most vendors update
computer and eradicates the their virus definition files on a monthly basis, and many provide options for
viruses it finds. free downloading of these files from the Internet. Others provide for auto-

at
matic emailing of updated files or push technology to help reduce the
administrative overhead associated with virus protection. Some vendors also
provide alert services to warn users of new viruses that arise between regu-

y
larly scheduled updates. You need to make sure that all equipment that is

op
running the anti-virus software receives updated copies of the definition files.
eM
4. Install the anti-virus software according to the manufacturer’s instructions.
Make sure that the virus
Most anti-virus software will prompt you during installation to create an
definition files you download emergency recovery or boot diskette. Make sure that you do this, and update

tC
and use have been or recreate the emergency diskette each time you update the virus signature
developed for the specific files. Although this might seem like a lot of work, if a virus ends up on one
version of the engine you are of your computers, using an up-to-date emergency diskette is probably the
using. For example, if your easiest way to recover from the virus infection.
home computer is using
No
version 3.x of a virus engine,
c

don’t mistakenly download

version 4.x definition files.
Scanning Options
en

Be aware that there are several ways to scan for viruses, including:
Do

• On-demand. On-demand scans can check a file, a directory, a drive, or a

whole computer for viruses. You can run on-demand scans either manually
Please note that these
options aren’t mutually or according to a schedule. On-demand scans provide a sort of preventive
exclusive. You can employ maintenance with regard to virus protection, but only if the virus signature
any combination of these files are current.
se
r

options to get the virus

protection you need for your
network.
fe
ea
Pl
Re

414
• On-access. On-access scans check only the files you’re accessing. They often
run in the background when you insert a floppy disk, download a file from
an ftp site, receive emails and attachments, run Java applets or ActiveX pro-
grams, or access Web pages. Although they can slow performance, on-access
scans can do a lot to prevent viruses from gaining access to your network.
• Heuristic. Some anti-virus software provides you with the option to perform

ial
heuristic scanning. Heuristic scanning goes one step beyond traditional scan- Some programs won’t allow
ning in that it enables an engine to search for viruses that aren’t listed in the you to install them if anti-
virus definition file by looking for and reporting unusual activity that might virus software is running. To
be a sign of virus infection, such as odd creation or modification dates. One be able to install the software
drawback to heuristic scanning that you should be aware of is that it can without risking virus
create numerous false alarms. infection, perform an on-
demand scan of the

er
• Local. Local scanning is by far the most prevalent in a network installation executable,
environment. With local scanning, the anti-virus software (engine and defini- disable the anti-virus
tion files) reside on each individual computer and is run from a local drive. software, install the new
application, and reactivate
• Online. Many vendors provide Web pages from which you can scan a com- the anti-virus software
puter for viruses. Often, additional files such as plug-ins need to be immediately after you

at
downloaded before you can perform an online scan. complete the installation.

y
What to Do When You Find a Virus

op
eM
If a virus is detected on a workstation computer, you should scan any hard or
floppy disks on that computer that might also be affected. You can quarantine the
affected area by asking users to stop using their computers until you’ve com-

tC
pleted the cleanup. Scan all computers in the quarantine area. For those
computers that don’t have the virus, verify that the virus signature files are up to
date. For those computers that have the virus, clean the virus and update the sig-
nature files. If any of the computers won’t start or run properly, use an
No
emergency anti-virus boot diskette before you clean them.
c

TASK 16C-1:
en
Do

Discussing virus protection

1. What are the two components of anti-virus software that you can update
to ensure a proper level of virus protection?
se
r

2. If you can’t save a Microsoft Word document, what’s the most likely
fe

cause and solution? Explain your response.

ea

a. Your computer is infected with a boot-sector virus. You’ll have to

format the hard drive and reinstall all software.
Pl
Re

b. Your computer is infected with a worm virus. You’ll have to scan

and clean all viruses from the system.
c. Your computer is infected with a boot-sector virus. You’ll have to
copy the file into WordPad and restart Word.

Lesson 16: Maintaining the Network 415

 d. Your computer is infected with a macro virus. You’ll have to scan
and clean all viruses from the system.

3. List at least two ways of obtaining updated virus signature files.

ial
4. A user downloads a freeware program from the Internet onto a Win-
dows NT Workstation. The next day, when he tries to start the
computer, the Missing Operating System message is displayed. This
computer has the latest virus scanning software and definition files
installed. You must get the computer working properly and get rid of
the virus. Optionally, you also want to ensure that the computer is pro-

er
tected from downloaded files and viruses. A colleague suggests that you
reboot the computer in Safe Mode, then run an on-demand scan and
clean all files, reconfigure the virus scanning software to scan and clean
all downloaded files, and reboot the computer.
a. The proposed solution meets the required results.

at
b. The proposed solution meets the required results and one of the
optional results.

y
c. The proposed solution meets the required results and both of the
optional results.

op
eM
d. The proposed solution doesn’t meet the required results.

tC
Suggested time:
APPLY YOUR KNOWLEDGE 16-1
30 minutes
Performing an online scan
No
In this activity, you’ll connect to an anti-virus Web site and perform an
c

on-demand scan for viruses. You’ll need an Internet connection, a Web browser
(Microsoft Internet Explorer 4.x or higher, or IE 4.x or higher and Netscape
Communicator or Navigator 4.x), and an email address to complete this activity.
en
Do

1. If necessary, start your Web browser, and close all other open programs.

2. Go to the McAfee Web site (www.mcafee.com) and enter the Anti-Virus

Center.

3. Click Scan Now. If you’re using Netscape Communicator or Navigator,

se
r

download the Netscape Activator plug-in. Acknowledge any security mes-

sages you receive. (You should acknowledge any security messages you
fe

receive throughout this activity.)

ea
Pl
Re

416
4. Review the Scan Now page and enter your email address.

5. Complete the registration form and click Free Trial.

6. Accept the license agreement.

7. Start the download of VirusScan Online components.

ial
8. Click the McAfee VirusScan Online link.

9. In the Scan In box, select your hard drive and click Scan.

10. Observe as the program files on your hard drive are scanned for viruses.

er
11. When the scan is complete, clean any viruses found.

12. Exit the browser.

at
TOPIC 16D

y
Patches and Other Software Updates

op
eM
Because no software package is perfect, manufacturers provide free updates, or
patches, to fix known problems or to provide enhanced services with the
software. They provide temporary solutions until a new version of the product is
released. Some patches can be applied ″on the fly″—that is, without shutting

tC
down the server; others require you to down the server to apply the patch patch:
correctly. A free update provided by a
software manufacturer that
Applying NOS patches can enable you to improve network performance without address known problems or
No
hardware upgrades or complete software upgrades. Although you might not need provides enhanced services.
c

(or want) to apply every patch released for your NOS, you should be aware of Also referred to as fixes,
the patches that are available and the issues that they address. You’ll get this support packs, and service
packs.
information automatically from the vendor if you have a monthly CD subscrip-
en

tion; otherwise, you’ll need to check the vendor’s Web site or subscribe to a
Do

mailing list to get this information. Major software vendors such as Microsoft and
Novell provide quarterly reports of the patches and fixes they’ve released. The
same is true for desktop operating systems and user applications—you should
monitor the patches released and determine which of them should be applied to
workstations—as well as for hardware-updated device drivers are often made
se

available by both hardware and software vendors.

r
fe

Patches
ea

Patches, also referred to as support packs or service packs, enable you to update
your operating system or client software without having to upgrade your system
Pl

completely.
Re

Lesson 16: Maintaining the Network 417

 Determining if Applying a Patch is Necessary
It would seem that if your NOS vendor has developed a patch to solve a particu-
lar problem, you ought to apply that patch automatically, but that’s not always the
case. Some patches can cause new problems while fixing the identified problem.
There are two basic approaches to patching NOSs. The first is to apply only those
patches necessary to keep the system up and running. In other words, don’t

ial
bother fixing what isn’t broken. The other approach is to keep the server software
as up-to-date as possible to take advantage of all possible features and enhance-
ments made to the software. You need to determine which approach best meets
the needs of your network. If network stability is the top priority, you’ll probably
go with the first approach; if you always need the most up-to-date features, you’ll
probably go with the second approach.

er
Where to Get Patches
You can get patches from several sources, including the vendor’s Web site, the

at
vendor’s support subscription service (CD or DVD), or an approved reseller. In
some cases, you’ll also get desktop operating system patches automatically when
you buy a new computer.

y
It is extremely important that you obtain patches only from a trusted source. Do

op
eM
not use patches from online magazines or shareware sites on a production server
or other computer.

tC
Applying and Testing Patches
Because applying a patch changes the network environment, you should carefully
plan before you apply any patch file. Whenever possible, you should test the
No
patch in a non-production environment to see how it works with other files. This
c

is especially important with patches that are still in testing. You can use the fol-
lowing general procedures to download and apply a patch file:
1. Familiarize yourself with the changes the patch is designed to provide, by
en

reviewing information on the manufacturer’s Web site or other available

Do

documentation.
2. If you determine that the patch file is suitable for implementation on your
network, download the patch and all related documentation to an isolated
test network or other isolated computer.
se

3. Access and carefully examine all documentation that accompanies the patch.
r

Readme files can provide specific implementation directions you need to

know before trying to apply the patch, as well as a list of the features that
fe

will be changed by applying the patch file.

ea

4. Create a checklist that includes each of the anticipated changes and new fea-
tures and a method to test them.
5. Start installing the patch on a test server or workstation, as appropriate.
Pl
Re

Make sure that you use an installation method that enables you to save pre-
vious configuration information, in case you have to uninstall the patch. If
this isn’t possible, create a backup of the test machine.
6. As you work through the installation, record all options, such as retaining or
replacing drivers and other files, and the selections you make.

418
7. When the installation is complete, restart the computer. If the computer
won’t boot up, repeat steps 5 through 7 with a clean test machine, but try
keeping the original device drivers. Do not proceed until you can complete
this step!
8. Use your testing checklist to verify that all features work as advertised.
9. Keep the test system running for at least two weeks, rebooting and trying

ial
different tasks to verify that all parts of the software continue to work
properly. Do not proceed until you can verify the stability of the patch in this
test environment.
10. If problems don’t arise during the test period, perform a limited deployment
of the patch. It’s a good idea to use IS support staff (and their servers and
workstations) as ″guinea pigs″ for the limited rollout, because they can fur-

er
ther test the patch, and if something does go wrong, they’ll probably be able
to recover more quickly that line workers would be able to. Do not proceed
until you can verify the stability of the patch in this test environment.
11. After the IS staff has tested the patch and determined that it’s safe to distrib-

at
ute, deploy the patch to a larger group of users’ workstations and servers. If
possible, this group should not be directly involved in the production
process. Do not proceed until you can verify the stability of the patch in this

y
test environment.
12. Use an automated rollout procedure to deploy the patch to the rest of the

op
network servers and workstations.
eM
13. After you have completed the rollout, verify that all equipment has the same
patch applied, and document the entire process as part of your network

tC
documentation and revision control measures.

Device Drivers
No
c

Device drivers act as interfaces between the network operating system and hard-
ware devices such as hard disks, CD-ROM drives, and network cards. On the
server, you might need to maintain several different types of device drivers.
en

Check the documentation for your NOS to determine the types of device drivers
Do

used on your network. To get the most current device drivers, check with the
hardware vendor as well as the software vendor.

Keeping Track of Patches Applied

se
r

As part of your network documentation scheme, you should record all patches
and other software updates applied to each computer. That way, if you have to
fe

rebuild the machine, you’ll be able to recreate the system accurately.

ea
Pl
Re

Lesson 16: Maintaining the Network 419

 TASK 16D-1:
Discussing patches and updates
1. Where’s the best place to get patches and updates for your NOS?

ial
2. Other than NOS patches and application updates, what other types of
files need to be monitored for updates?

er
3. You have to install a patch for the spreadsheet application used by the
entire company. In what order should this patch be applied? Explain
your response.
a. A test machine, your workstation, management workstations, pro-

at
duction workstations.
b. A test machine, your workstation, support staff workstations,
selected users’ workstations, production workstations.

y
c. Your workstation, selected users’ workstations, production

op
eM
workstations.
d. A test machine, your workstation, selected users’ workstations, pro-
duction workstations.

tC
4. How often should you check for updates to network operating systems
and applications?
No
c

5. How long should you test an OS patch before rolling it out to produc-
tion servers?
en
Do

Suggested time:
APPLY YOUR KNOWLEDGE 16-2
30 minutes
Maintaining the network
se
r

In this activity, you’ll work through some sample test questions that deal with the
information presented in this lesson.
fe

1. Load the Skill Assessment component of the Network+ Certification

ea

CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

Pl
Re

3. Select all items for the Maintaining and Supporting the Network topic,
uncheck the Test Time check box, and then click OK.

4. Work through the sample questions and score the test.

420
5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

6. Exit the test.

ial
Summary
In this lesson, you’ve identified sources of network documentation, backup
strategies, virus-protection strategies, and the importance of patching operat-

er
ing system and applications.

LESSON 16 REVIEW

at
16A What is the best source of free information about a networking prod-

y
uct?

op
eM
16B List some information that should be included in a backup plan.

tC
No
c

16CWhere’s the best place to install anti-virus software?

en
Do

How often should you update virus definition files?

16D How can you determine which NOS patches you should apply to your
se
r

network?
fe
ea
Pl
Re

Lesson 16: Maintaining the Network 421

Re
fe
Pl
r en
ea c
se eM
Do
No at
tC er
op
y ial
Troubleshooting the LESSON
Network
17

ial
Data Files:
Network+ Certification
Overview CD-ROM

er
This lesson describes the basics of network troubleshooting practices. Lesson Time:
3 hours

Objectives

at
To be able to provide service and support to existing networking implementa-
tions, you will:

y
17A Explore troubleshooting models.

op
eM
This topic describes a systematic approach to solving network problems.

17B Identify troubleshooting tools.

tC
In this topic, you’ll discuss some hardware and software tools that can
help you diagnose and correct network problems.
No
c
en
Do
se
r
fe
ea
Pl
Re

Lesson 17: Troubleshooting the Network 423

 TOPIC 17A
Exploring Troubleshooting Models

ial
Using a Systematic Approach to Troubleshooting
You can define troubleshooting as the timely restoration of network services.
However, you must ensure that affected users agree that network services have
been properly restored before you close a request for service. Consider a problem
solved only when that problem no longer has an adverse effect on users.

er
Troubleshooting is seldom easy, but if you develop a consistent plan of attack,
you will often be able to determine the specific cause of the problem, as well as
possible solutions to the problem. Experience can play a key role, enabling you
to analyze a user’s problem (such as ″I can’t print to the printer down the hall″)

at
and determine what part of the network is causing the problem (such as the
cabling, the printer, the workstation hardware, or the workstation client software).

y
First Things First

op
eM
Most times, you’ll be informed of network problems by users who can’t perform
some function that they expect to be able to do. In other cases, you might dis-
cover the problem on your own or be alerted to the problem by network

tC
management software. No matter where the indication of the problem comes
from, there are a few things you can do to narrow down the possible cause of the
problem.
• Check for simple or obvious things. To paraphrase Occam’s Razor, all things
No
being equal, the simplest solution to a problem is often the correct one.
c

— If a user reports a problem such as not being able to access the net-
work, make sure that he or she is using the correct login procedure,
login name, and password. In case-sensitive environments, such as
en
Do

UNIX, verify that the user’s [Caps Lock] isn’t active. Also, verify that
the user has rights to access the network at the day and time of the
problem, since in NetWare and Windows NT environments, system
administrators can limit login days and times, as well as the number of
simultaneous login sessions a user can have.
se
r

— User error is a common reason for errors. Eliminating user error before
investigating other causes can solve the problem much of the time.
Make sure that the user correctly follows the procedure for doing the
fe

task in question. Sometimes, the user follows the procedure and it

ea

works, but the user expected different results and thinks that the proce-
dure has failed. In a situation like this, explain the expected results to
the user to avoid confusion in the future. If the user can’t access files or
Pl
Re

applications on the network, make sure that the correct rights have been
assigned. Sometimes the user is expecting to be able to change data in a
file, and has been granted only enough rights to view the contents of
the file. Check to make sure which set of rights the user should have
been given. Another common reason users can’t access files or applica-
tions is that their login failed (perhaps they entered the wrong user

424
 name or password). If the user doesn’t realize that login has failed and
tries to perform a task, they won’t be able to do it.
— Check that all components are in place, that they are the right ones, and
that they’re connected correctly. A loose connection can cause unex-
pected results. Another common scenario is that a user tried to install a
new component and failed to do so correctly.

ial
— If the problem seems to be related to a hard drive, make a backup of
the data on the drive, and test the backup to ensure that it is valid ,
before you proceed further with troubleshooting the problem.
— Sometimes, simply rebooting the user’s workstation will fix the
problem. If this happens frequently, you’ll need to look further into the

er
cause. At other times, the user might not have told you all of the
actions performed before the computer stopped working correctly, so
rebooting resets the machine so that all default parameters are back in
effect.
• Check for trouble indicators.

at
— Power lights and switches can provide a fast solution to some user
problems. For instance, you might receive a service request to fix a
monitor that doesn’t come on when the user powers up the workstation.

y
Verifying that the monitor is turned on and plugged in can save you
time you might have otherwise invested in researching the problem.

op
eM
You should also check all other power cables and power strips, working
your way back from the device to the power service panel and checking
all possible causes such as power cables, outlets, electrical wires, circuit

tC
breakers, and fuses.
— If network connectivity is in question, check the link lights on a NIC
and a hub to verify that the client software (or NIC driver) is loaded
and that Data Link layer communication is happening between the
No
workstation and the hub.
c

— Error messages and log files can sometimes provide a starting point for
determining the cause of a problem. If a user calls with a problem, have
them tell you exactly what the error message stated. If the error mes-
en

sage is no longer displayed on their workstation, have them try and

Do

re-create the problem so that they can tell you what the error message
says. You can also check server consoles and log files for error mes-
sages when a problem seems to be widespread.
— Performance monitors and network protocol analyzers can also help
pinpoint the source of a problem. For instance, if users report slow net-
se
r

work response times after 4 pm every day, check for increased network
utilization during that time period. If your database server shows high
fe

utilization, perhaps the Accounting department runs daily reports during

ea

that time.
• Determine if the problem is attributable to hardware or software.
— Hardware problems often materialize as a device not operating and an
Pl
Re

error message being displayed when you try to use the device. Some-
times you’ll get warning signs, such as a Disk I/O error on a hard disk
that is close to failing; other times, the device will simply stop working.
Solving hardware problems usually requires that you change hardware
settings when a conflict exists or replacing the hardware when there
aren’t any conflicts. Because the downtime for hardware failures can be

Lesson 17: Troubleshooting the Network 425

 significant, especially if you need to send the unit out for servicing, it’s
a good idea to have backup or replacement hardware on hand.
— Software problems can manifest itself in a number of ways. Error mes-
sages might or might not be displayed, and those displayed might or
might not be helpful. Or an application or even the entire machine
might lock up, requiring you to close the offending application or

ial
reboot the computer. If error messages list missing or corrupted files,
you can replace the files or reinstall the application; otherwise, you’ll
probably have to visit the manufacturer’s support Web site and see if
this problem has been reported and if any patches or fixes are available.
Another thing to remember is that some intermittent software problems
can be caused by fragmented memory, which can be caused by opening

er
and closing many files and applications. Try closing all applications and
rebooting the computer—sometimes the solution is that easy.
• Determine if the problem is limited to one workstation, several workstations,
one server, one segment, or the entire network. If only one person is experi-
encing a certain problem, the problem is most likely at the workstation. If

at
groups of users are affected, the problem might lie at a server or some other
part of the network that the users all have in common.

y
— When one user reports a problem, such as not being able to log in to
the network, try logging in as that user from another workstation in the

op
eM
same group of users. If login is successful, start by checking the work-
station’s NIC and cabling, then move on to more detailed workstation
troubleshooting.
— When several users report the same problem, find out what they all

tC
have in common. For instance, if all the affected users use the same
server, verify that the server is up and running smoothly, and check the
user connections and security levels. If several network segments appear
to be affected, check for network address conflicts. If all users are hav-
No
ing problems, check any components (such as servers, routers, and
c

hubs) that all users access. Also remember to check any WAN connec-
tions by verifying that stations on both sides of the WAN link can
communicate; if they can’t, you’ll need to check the WAN hardware
en

along with other devices between the sending and receiving stations.
Do

• Check cabling. After you’ve determined the extent of the network problem,
verify that all cables in the problem area are connected to the proper ports
and that they are working properly (link lights should be on). Also, verify
that workgroup hubs are properly connected and configured.
se
r

Troubleshooting Models
fe
ea

Because troubleshooting network problems is such a big part of a network admin-

istrator’s or network engineer’s job, there are many troubleshooting models to
choose from. The important thing to remember is that you should always use
Pl

some sort of systematic approach to problem-solving.

Re

Troubleshooting can be a difficult process. It’s unlikely that anyone can develop a
complete and accurate ″cookbook″ approach to troubleshooting, because trouble-
shooting is often done through intuitive guesses based on experience.
Troubleshooting models simply attempt to create processes that help when you

426
have no experience on which to base your troubleshooting techniques. Remember
that even experienced troubleshooters can have blind spots that allow them to
overlook obvious problems or solutions. Using a troubleshooting model helps
prevent a blind spot from interfering with your ability to find a solution to your
problem.
Before troubles arise, you can take some preventative measures, such as training

ial
users, keeping an inventory of spare working parts, and maintaining valid back-
ups of important data.

Network+ Troubleshooting Model

er
The troubleshooting model covered in the Network+ exam is an 8-step process.
1. Identify the exact issue. Ask the user a series of open-ended questions to
help identify the issue behind the symptoms. For instance, instead of asking
if a user can log in to the network, try asking what happens when the user
tries to log in.

at
2. Re-create the problem. If possible, have the user try the procedure again,
recording all actions taken and all results received. The best case is to watch

y
the user to verify that he or she is following the correct procedure. If the
problem occurs in the same place, the problem will be easier to solve than if

op
eM
it’s an intermittent one. When possible, also try and re-create the problem in
the following ways:
• Try to perform the task yourself at the user’s workstation and at your
workstation.

tC
• Have another user try the task at the user’s workstation and on an
equivalent workstation.
3. Isolate the cause. After you reproduce the problem, try to determine what’s
No
causing it. Use a systematic approach to eliminate possible causes, starting
c

with the most obvious cause and working back through other causes. You
can also ask the user (and yourself) questions similar to the following to
help isolate the cause:
en

• Could you do this task before? If not, maybe the system is simply
Do

unable to perform the task without additional hardware or software.

• If you could do the task before, when did you first notice that you
couldn’t do it anymore? If the computer suddenly stops doing what it’s
always done seamlessly, that might not be the only change. Try to dis-
cover what happened immediately before the problem arose, since the
se
r

source of the problem might be related to other changes.

• What’s changed since the last time you were able to do this task? Users
fe

might tell you that nothing’s changed, so you should follow up with
ea

leading questions such as ″did someone add something to the com-

puter?″ and ″is the procedure you followed this time different in any
way from the way you normally do this?″
Pl
Re

• Were error messages displayed? If you can get the exact text of any
error messages displayed, you can try searching the manufacturer’s Web
support site to get an explanation of the message and to see if any
problem reports have been logged related to the message.
• Is the problem always the same, no matter what conditions apply?
Determining if the problem is consistent or intermittent can help you

Lesson 17: Troubleshooting the Network 427

 narrow down possible causes. For instance, if a user can’t open a
spreadsheet file stored on a network drive, ask if they can open a differ-
ent spreadsheet file stored on the network, or a spreadsheet file stored
on the hard drive. If these files open successfully, you can be pretty
sure that the problem lies with the file, not the application.
4. Formulate a correction. Depending on the cause you’ve isolated, determine

ial
at least one way to correct the problem. For example, if you’ve isolated the
cause as a corrupt spreadsheet file, one correction would be to restore the
file from a backup. Draw on your own experience, review support Web sites,
and confer with your colleagues to come up with possible corrections. And if
you come up with more than one correction, prioritize them according to
their likelihood of success and ease of implementation.

er
5. Implement the correction. This step is where you actually fix the problem.
Before you do so, however, you need to ensure that productivity doesn’t suf-
fer and that downtime is minimized. For example, you might need to
provide a ″loaner″ workstation to a user whose machine needs to be rebuilt.
6. Test the solution. Make sure that the solution that you’ve implemented actu-

at
ally solved the problem and didn’t cause any new ones. Use several options
and situations to conduct your tests; for instance, try the task yourself, then

y
have the user try the task while you observe the process, or test the worksta-
tion both before and after it’s connected to the network. Sometimes you’ll

op
eM
need to test over time to ensure that your solution is the correct one.
Remember to verify that the user agrees that the problem is solved before
you proceed.
7. Document the problem and the solution. It’s a good idea to create detailed

tC
If testing reveals that your
descriptions of network problems and their solutions, and maintain them as
solution doesn’t solve the part of your overall network documentation plan. Not only will this provide
problem, or it causes new you with an ever-growing database of information specific to your network,
problems, you’ll need to it will also be valuable reference material for use in future troubleshooting
No
backtrack and repeat steps 4 instances. You might even want to create a troubleshooting template so that
c

through 6. Be sure to record you can be sure that necessary information is included in all trouble reports,
anything that doesn’t work so
that you don’t waste time and that all reports are consistent, no matter which support person completes
duplicating the same results. them. Some of the things you’ll want to include in a troubleshooting tem-
en

plate include:
Do

• A description of the initial trouble call, including date, time, who’s

experiencing the problem, and who’s reporting the problem.
• A description of the conditions surrounding the problem, including the
The person experiencing the
type of computer, the type of NIC, any peripherals, the desktop operat-
se
r

problem isn’t always the one ing system and version, the network operating system and version, the
to actually report the version of any applications mentioned in the problem report, and
problem. A coworker, whether or not the user was logged in when the problem occurred.
fe

manager, or even Help Desk

•
ea

or other support personnel The exact issue you identified in step 1 of this process.
might report a problem for a • Whether or not you could reproduce the problem consistently.
user.
• The possible cause or causes you isolated in step 3 of this process.
Pl
Re

• The correction or corrections you formulated in step 4 of this process.

• The results of implementing each correction you try.
• The results of testing each solution.

428
 • Any external resources you used, such as vendor documentation,
addresses for vendor and other support Web sites, names and phone
numbers for support personnel, and names and phone numbers for third-
party service providers.
8. Provide feedback. Don’t underestimate the importance of this step. You
should never consider a problem to be resolved until the customer considers

ial
it to be solved. You’ll probably also need to inform others of the outcome of
the situation, especially in instances where the person reporting the problem
is not actually the person experiencing the problem. When you can, provide
a brief explanation of the problem and how you fixed it, but make sure that
you don’t overwhelm the user with information, and never blame the user
directly for the problem. Your explanation should always be geared to the

er
knowledge and interest level of the person you’re addressing. Consider a
user who can’t access the local external drive she uses with her laptop. You
receive the problem report from a Help Desk technician. After working
through the troubleshooting process, you determine that the problem is a bad
PCMCIA card, which was probably fried when the user inserted the card

at
while the laptop was running. What you might tell the user is that a hard-
ware component went bad and that you’ve replaced it; then, without being
accusatory, review the proper procedures for removing and inserting

y
PCMCIA cards, and demonstrate that the external drive is available again.
What you might tell the Help Desk technician is that you replaced a bad

op
eM
PCMCIA card, which probably got fried due to the user not inserting the
card before booting the laptop, and that you’ve shown the user the correct
procedure for working with PCMCIA cards.

Novell Troubleshooting Model

Another troubleshooting model offered by Novell is a 6-step process.
tC
No
c
1. Try some quick fixes. Before getting too deep into the troubleshooting mode,
you should check some of the obvious causes of errors or problems. Obvious
causes are frequently overlooked, causing you to waste time searching for
en

more complicated reasons for the trouble.

Do

2. Gather basic information. This step involves both gathering information

before trouble happens, and gathering information about trouble that is
occurring. Use the following guidelines to help you complete this step:
• Determine the symptoms of the problem, what users are affected, and
under what conditions the symptoms occur.
se
r

• Determine the amount of network traffic when the problem occurs to

ensure that high traffic isn’t the problem.
fe

• Compile baseline data for network performance and operating

ea

conditions. Compare the current trouble with the baseline data to deter-
mine if discrepancies are present.
• Find out if this computer worked previously, and determine what’s
Pl
Re

changed since it last worked correctly.

• Check user and usenet groups, as well as online forums, to see if others
have encountered similar problems.
3. Develop a plan of attack to isolate the problem. In this step, you determine
how you’ll isolate and correct the problem. Use the following guidelines to
help you complete this step:

Lesson 17: Troubleshooting the Network 429

 • Use the basic information that you’ve gathered and your background
knowledge to determine two or three possible causes for the problem.
Possible causes might be user error, application error, operating system
error, or hardware error.
• Prioritize the likely causes of the problem and your solutions to them.
You can use the following factors to help you prioritize: relative ease of

ial
solution, time required to implement solution, likelihood of success, and
cost of solution.
• Document your strategy for later reference. If your plan works this
time, it might work later; if it didn’t work, you might want to develop a
different plan the next time you have a similar problem.

er
4. Execute the plan. This step begins the actual troubleshooting operation. The
goal of this step is to find the cause of the trouble. Use the following guide-
lines to help you complete this step:
• Analyze the first possible cause to determine the smallest testable steps.
• Make one change at a time, and test the change to see if it solves the

at
problem. If the change doesn’t solve the problem, undo it. Document all
changes as you proceed so that you have a record of what you have
tried, in case you need to backtrack.

y
• Use a forward-chaining, backward-chaining, or binary-chaining method

op
eM
to uncover the source of the trouble. With forward chaining, you start
from the server or other source device and work toward the
workstation. With backward chaining, you start from the workstation
and work toward the source device. With binary chaining, you start

tC
halfway between the source device and the workstation and determine
which half holds the problem. Divide that segment in half, and deter-
mine which half holds the problem, repeating until the source of the
problem is isolated. By the end of this step, you should have solved the
No
problem. If the problem still persists, your starting assumptions might
c

have been incorrect. Repeat this step by considering another possible

cause of the problem.
5. Verify user satisfaction. Remember, you should never consider a network
en

problem to be solved until the users consider the problem solved. When nec-
Do

essary, train users on how that part of the network should work. Also, you
might need to wait until the solution has been in place for a while before
you can consider the problem to be solved.
6. Document the problem and the solution. Documentation is the key to solving
future problems, getting equipment upgrades, and preventing financial losses
se
r

from network troubles. Use the following guidelines to help you complete
this step:
fe

• Record the exact nature of the problem in a log book, along with the
ea

solution and the method you followed to reach the solution.

• Make plans or changes that will prevent future occurrences of this or
similar problems. For example, if the failure was due to a faulty BNC
Pl
Re

connection on a coaxial cable, you might want to use different connec-

tors, use different installation procedures, or switch to another cable and
connection type, such as twisted pair.
• Regularly scan the network for viruses by using up-to-date virus scan-
ning software.

430
 • Monitor network performance by using current network management
software.
• Implement a plan to regularly test the system, even when no problems
are apparent. Document the tests and the circumstances surrounding
them.
• Develop and maintain a network map and floor plan as part of the net-

ial
work documentation. These can help you spot potential trouble spots.

Collect, Isolate, and Correct Model

A third popular troubleshooting model is the Correct, Isolate, and Correct method.

er
The Collect, Isolate, and Correct method divides the troubleshooting process into
three large stages, each containing several steps. Ultimately, it achieves the same
goal as other troubleshooting methods, which is to solve network problems.
1. Collect. In this stage, you gather information, gather user reports, document
the process, and keep track of known problems. The steps involved in the

at
Collect stage might include:
• Document the troubleshooting process.

y
• Gather user reports of the trouble.
•

op
eM
Gather error messages and view the relevant error logs.
• Make sure that the user is following proper procedures. You might want
to check the order that the user performs certain steps, like loading soft-
ware, logging in upon starting Windows, and so forth.

tC
• Perform software diagnostics. This step can be performed remotely,
before you go to the user’s site.
• Gather the necessary test equipment and tools to take to the customer’s
No
site.
c

• Gather working replacement parts. Test these parts to ensure that they
work. Have them available to use when problems appear. User and customer can be
• Monitor the performance of your network and develop baseline used interchangeably since
en

statistics. Compare current conditions to your baseline statistics and pre- you should consider all
Do

pare for problems when parameters greatly exceed your baselines. users your customers and
treat them accordingly. In
• Track known problems and bugs by consulting with other experts and some cases, users will
with online sources, and by reading periodicals and trade magazines. literally be customers (if you
Attend local users’ groups to stay informed. are a contractor or service
provider).
se

•
r

Maintain an accurate log of the configurations of your equipment, your

network, and your users’ environments.
fe

The key to this stage is to gather a sufficient amount of high-quality

ea

information, rather than simply a large quantity of information. The

experience you gain from troubleshooting your systems will help you
determine what data you need to maintain.
Pl

Some of the steps in the Collect stage are done long before trouble
Re

arises on your network. Being prepared is the best way to solve prob-
lems quickly and easily when they do appear.
2. Isolate. This stage relies on the proper completion of Collect stage items.
The steps involved in the Isolate stage might include:

Lesson 17: Troubleshooting the Network 431

 • Plan your attack on the trouble and document your chosen method. Fol-
low your plan.
• Follow a hierarchical troubleshooting process. Start with the big picture.
Check to see how many users are affected by the same problem. Start
with the server, or those items that affect all of the users experiencing
problems. After eliminating the server, check the network or intervening

ial
networking devices. Don’t waste time checking each individual station
if many users are affected; the problem is unlikely to be at the end-
users’ stations.
• Start with the easiest problem first. Check to see that the power cord is
plugged in before you check to see if the internal power supply is
burned out. Check for obvious and logical problems before looking for

er
strange or unlikely problems.
• Follow a sequential method to troubleshoot. Complete one testing or
isolating process before beginning another.
• Document the steps you follow to isolate the problem. Keep track of

at
any changes you need or want to make before you implement them.
• Ask isolating questions to narrow the range of possibilities for the cause
of the problem.

y
• Make sure that proper versions of software and hardware are being

op
eM
used. Keep a detailed record of versions, upgrades, and known
problems. This record will help you troubleshoot problems.
• Swap parts as necessary to determine the cause of the problem. This
will help you to find faulty components quickly and easily. However,

tC
swapping parts won’t necessarily help you find problems arising from
the interaction between two or more parts that you are swapping.
• Challenge your own assumptions. Or, better yet, don’t make assump-
tions when you are troubleshooting. If you find yourself saying ″of
No
course, that can’t be the problem,″ you might need to challenge your
c

assumption and try changing that anyway. This is especially true when
you seem to have exhausted all other possibilities.
en

The Isolate stage is a balancing act between a methodical series of steps

Do

and a best-guess attempt at solving problems. In some cases, an intui-

tive reaction will lead you quickly to the source of the trouble. In other
cases, you’ll need to methodically proceed with the troubleshooting pro-
cess to discover the root of the problem.
The Isolate stage is often completed nearly simultaneously with the
se
r

Correct stage.
3. Correct. This stage involves the steps that actually fix the problem, as well
fe

as preventive measures. The steps involved in the Correct stage might

ea

include:
• Document any attempts you make to solve the problem. Note the condi-
tions before and after you’ve implemented your correction.
Pl
Re

• Take one step at a time. Make only one change; then test to see if the
problem has been solved. If so, you’ve found the problem. Document
your solution and take the necessary steps to complete the trouble call.
However, if your step didn’t fix the problem, undo your changes. Make
a change only if you can undo it.

432
 • Test your final solution for full functionality. Ask the end-users to test
as well. Their perception of ″fully functional″ may not match yours. It
is better to discover further problems immediately than to get another
trouble call.
• Take any steps you can to prevent future troubles. For example, if you
noted a potentially faulty coax connector on the station cable, even if

ial
that did not turn out to be the cause of the trouble, you’d be wise to fix
it while you are at the user’s site. This will save you the time of return-
ing later to fix a problem you were already aware of.
• Check to see if others have a similar problem. For example, if the solu-
tion to this problem was to update the version of a driver on the user’s
system, check to see if everybody is in need of the upgrade. Your log

er
will help facilitate this step.
• Document the problem and its solution. This will aid you in trouble-
shooting future similar problems.
• Develop standards to prevent problems. For example, develop a stan-

at
dard file-server naming scheme so that when a new server is added, it’s
not given the same name as an existing server.
• Determine if this problem is made worse by your disaster recovery

y
plan. For example, if solving this problem involves restoring data from
tape, was the backup data available? Should your disaster recovery plan

op
eM
be modified to prevent similar problems in the future? If so, fix and
document your disaster recovery plan.
The prioritized list from the Isolate stage help you set a plan for solving

tC
the trouble. Follow your plan. Proceed down your list of potential
causes, step by step, from top to bottom. Document your progress, and
don’t skip any steps.
Ultimately, the troubleshooting process that you follow will be a mix of
No
these steps, plus methods of your own. Remember that a logical,
c

methodical approach to troubleshooting usually leads to quicker

solutions. Sufficient and proper documentation of setups, configurations,
topologies, and histories proves invaluable to the troubleshooting
en

process.
Do

TASK 17A-1:
se

Discussing troubleshooting models

r
fe

1. How can you determine the scope of a reported network problem?

ea

2. After you isolate the cause of a network problem, what’s the next step?
Pl
Re

3. Why is it important to document network problems?

Lesson 17: Troubleshooting the Network 433

 4. What are some quick fixes you can try when a user complains that she
can’t connect to the network?

ial
5. What’s the most often overlooked part of a troubleshooting plan?

6. What do all of the models discussed have in common?

er
7. Users on the third floor report that they can’t connect to the Internet,
but they can connect to the NetWare server. What would you check first,

at
and why?
a. Router configuration tables.

y
b. If viruses exist on the network.

op
eM
c. Power cable to the hub.
d. If users on other floors are having similar problems.

tC
Suggested time:
APPLY YOUR KNOWLEDGE 17-1
30 minutes
Using a systematic approach to troubleshooting
No
In this activity, you’ll use the Network+ Certification CD-ROM to work through
c

some troubleshooting scenarios.

1. Load the Activities component of the Network+ Certification CD-ROM.

en
Do

2. Click on the Network Protocols link, and then the Name Resolution link.

3. Point to the Resolve A Host Name Resolution Problem link. Read the sce-
nario, and then click on the link.

4. Complete the steps in the Activity as prompted in the lower-left corner of

se
r

the screen.
fe
ea
Pl
Re

434
5. When you’ve completed this Activity, complete at least one more Activity
from the following list. Work through as many Activities as time permits.
Network Protocols
Name Resolution
Resolve a WINS name resolution problem.

ial
Troubleshooting Tools
Use NETSTAT to troubleshoot a problem.
Use TELNET to troubleshoot a problem.
Network Operating System
Ongoing Maintenance with Windows NT

er
Use Event Viewer to troubleshoot a device failure.
Use WinMSD to troubleshoot a problem.

6. Exit the Activities component.

at
TOPIC 17B

y
op
Identifying Troubleshooting Tools
eM
In addition to using a systematic troubleshooting procedure, there are many other
tools and resources that you can use to streamline the troubleshooting process.

Hardware Tools tC
No
Hardware troubleshooting tools can help detect many Physical layer problems.
c

Some common hardware troubleshooting tools include crossover cables, hardware

loopback plugs, tone generators, tone locators, and cable testers.
en
Do

Crossover Cables
A crossover cable is a special network cable used in Ethernet UTP installations
that enables you to connect two hubs, or to connect two stations without using a
hub. In a crossover cable, the transmit and receive lines are crossed, a function
se

that is normally taken care of by the hub or the switch. In a regular Ethernet UTP crossover cable:
r

patch cable, four wires are used—two (Pins 1 and 2) to transmit and two (Pins 3 A special network cable used
and 6) to receive)—and all lines are straight-wired (Pin 1 is wired to Pin 1, Pin 2 in Ethernet UTP installations
fe

that enables you to connect

to Pin 2, and so forth). In a crossover cable, Pins 1 and 2 connect to Pins 3 and two hubs, or to connect two
ea

6, and Pins 3 and 6 connect to Pins 1 and 2. stations without using a hub.
Pl
Re

Lesson 17: Troubleshooting the Network 435

 If you connect hubs via a crossover cable, you can cascade the hubs to provide
more ports for a workgroup area, rather than buying and installing a larger hub.
In troubleshooting, crossover cables let you connect two workstations (or a work-
From the outside, patch station and a server) directly so that you can test communications between them.
cables and crossover cables For instance, if you suspect that a server’s NIC might be bad, you can use a
look the same, so make sure
crossover cable to attach a laptop’s NIC directly to the server’s NIC. Provided
you label crossover cables
that both NICs are configured correctly, you should be able to log in to the server

ial
as such. If you try to use a
crossover cable instead of a if the server’s NIC is good.
patch cable for a regular
workstation-to-hub
connection, the workstation
won’t be able to Hardware Loopback Plugs
communicate with the hub or
A hardware loopback plug is a special connector used for diagnosing transmis-

er
with the rest of the network.
sion problems. It plugs into a port and crosses over the transmit line to the
receive line so that outgoing signals can be redirected back into the computer for
testing. Hardware loopback plugs are commonly used to test Ethernet NICs. It
hardware loopback plug: directly connects the transmit pins to the receive pins, with Pin 1 being connected
A special connector used for to Pin 3 and Pin 2 being connected to Pin 6. If a NIC comes with hardware diag-

at
diagnosing transmission nostic capabilities, the loopback plug will be included with the NIC. Connect the
problems. It plugs into a port loopback plug to the installed NIC’s RJ-45 connector, and run the diagnostic soft-
and crosses over the ware to verify that the NIC can send and receive data.

y
transmit lines to the receive
lines so that outgoing

op
signals can be redirected
back into the computer for
eM
Tone Generator and Tone Locator
testing.
These pieces of equipment are most commonly used on telephone systems to

tC
trace wire pairs. Don’t confuse them with cable testers—they can only help you
determine which UTP cable is which.
A tone generator is an electronic device that sends an electrical signal through
one set of UTP wires. Tone generators are sometimes referred to as foxes.
No
A tone locator is an electronic device that emits an audible tone when it detects a
c

tone generator:
An electronic device that signal in a set of wires. Tone locators are sometimes referred to as hounds.
sends an electrical signal
through one set of UTP
To trace one cable in a group of cables, connect the tone generator to the copper
en

wires. Sometimes referred to ends of the wire pair you want to find; then move the tone locator over the group
Do

as a fox. of cables. A soft beeping tone indicates that you are close to the correct wire set;
when the beeping is loudest, you’ve found the cable. Don’t ever connect a tone
generator to a cable that’s connected to a NIC or a hub. The signal sent by the
tone generator can destroy network equipment.
tone locator:
An electronic device that
se
r

emits an audible tone when

it detects a signal in a set of Cable Testers
fe

wires. Sometimes referred to

as a hound. A cable tester can help you trace the source of cable problems on your network,
ea

such as short circuits, broken cables, cable lengths that are too long, bad connec-
tors, and high levels of interference.
Pl
Re

Software Tools
Many software-based troubleshooting tools are available on the market, and
through freeware and shareware conduits. Some of the most common types of
tools are protocol analyzers and performance monitors.

436
Virus scanning software can also be considered a troubleshooting tool. Even if
you have a good virus-protection plan implemented, remember to check for virus
infestations if the symptoms warrant it, or if the problem is unusual and can’t be
reproduced consistently.

ial
Protocol Analyzers
A protocol analyzer is a type of diagnostic software that can examine and display
data packets that are being transmitted over a network. They can examine packets
from protocols that operate in the Physical, Data Link, Network, and Transport
layers of the OSI model. Many protocol analyzers ranging in functionality are protocol analyzer:
available in the marketplace. Some have only software components; others use a Diagnostic software that can

er
combination of hardware and software to gather and analyze network information. examine and display data
packets that are being
Higher-end solutions usually provide support for more protocols, the ability to transmitted over a network.
send test traffic, higher speeds, and more analytical information than do the
lower-end solutions. The product that you should use depends on your specific
environment and the needs of your company.

at
Most protocol analyzers can capture and decode packets. When packets are cap-
tured, the protocol analyzer places copies of a series of packets into memory,

y
where the packets can be analyzed without disrupting network traffic. Decoding a
packet provides you with information about the source of the packet, the destina-

op
eM
tion of the packet, the protocol used to send the packet, and the protocol that
should receive the packet. From this information, you can determine the nature of
the traffic on the network, the protocols that are used most often, the types of
Web sites users are accessing, and which NICs are sending out corrupted or

tC
unnecessary packets. (If a NIC sends packets when it doesn’t have any data to
send, the NIC is said to be jabbering .)
Protocol analyzers require a special type of NIC and driver that supports promis-
cuous mode operation, enabling the station running the analyzer to recognize all
No
jabber:
packets being sent over the network, whether they are good or bad, and no matter A term describing a
c

what the source or destination is. In promiscuous mode, a network card passes all condition where a NIC sends
network events to the operating system and running applications. In normal unnecessary packets onto the
modes of operation, some network events are filtered out and not available to network, causing excessive
en

applications, including the error conditions that the protocol analyzer is designed traffic.
Do

to detect. Promiscuous mode drivers are available for most network interface
cards; contact the manufacturer of your cards for the most recent drivers.
Some of the more popular protocol analyzers available include Sniffer, Mentor, promiscuous mode:
NetLens, LANdecoder, Observer, Network Advisor, NetSight/LANDesk Analysis, A condition where a NIC
and LANalyzer. LANalyzer is available in two forms—the NetWare LANalyzer passes all network events up
se
r

Agent for ManageWise, and the standalone LANalyzer for Windows. the protocol stack to the
operating system and
running applications.
fe
ea

Performance Monitors
Many network operating systems include performance-monitoring software.
NetWare’s MONITOR.NLM utility and Windows NT’s Performance Monitor pro-
Pl
Re

vide information about the number of packets being sent and received, the
processing load of the server (server utilization statistics), and other performance
statistics that can help you pinpoint the source of bottlenecks and other network
performance problems.

Lesson 17: Troubleshooting the Network 437

 Third-party network performance monitors are also available, such as Tivoli’s
NetView and Lucent’s VitalSuite. Some protocol analyzers also provide
performance-monitoring capabilities.

Other Resources

ial
Other resources that you can use in your troubleshooting efforts include server
log files and other manufacturer resources such as readme files, phone-based sup-
port, support CDs, and support Web sites.

er
NetWare Log Files
NetWare log files include the console log (SYS:\ETC\CONSOLE.LOG), the
abend log (SYS:\SYSTEM\ABEND.LOG), the server and volume logs (SYS:\
SYSTEM\SYS$LOG.ERR and SYS:\SYSTEM\VOL$LOG.ERR).
•

at
The console log is a copy of everything—error messages and other system
information—that has been displayed on the server console during that
session. Console logging isn’t enabled by default; to enable it, use the

y
LOAD CONLOG command either at the server console or in the
AUTOEXEC.NCF file. When CONLOG is loaded, it overwrites the previous

op
eM
log file. To prevent the old console log from being overwritten, use the
Archive parameter when you enter the command (LOAD CONLOG
ARCHIVE=YES).
• The abend log is a recording of any abnormal ends, or abends, on a

tC
NetWare server, along with other information that can help you determine
the cause of the abend. Abends are error conditions that prevent the server
from operating properly. Abends can range in severity from one NLM or
service being unavailable to shutting down the entire server. Sometimes
No
abend: abends cause the server to reboot, which would mean that any console mes-
c

Short for ″abnormal end.″ A sages displayed immediately prior to the abend would be lost. NetWare 4.11
condition that prevents a
NetWare server from working and higher versions include a routine to save the output (messages) of the
properly. abend to the abend log file and to the server console.
en

•
Do

The server log file is a text file that contains all system messages and alerts
that are displayed on the server console.
You can tell if a server has
abended by looking at the
console prompt. The number
se

of abends since the server

r

was brought up is included

in brackets in the command
fe

prompt; for example,

ea

NWSVR1<2>:.
Pl
Re

438
• The volume log file is a text file that contains all messages and alerts per-
taining to the server volume. You should view these files periodically to see
what kinds of errors are occurring on your server. NetWare security viola-
tions are also recorded in the SYS$LOG.ERR file. Check this file daily if
you are concerned about security at your site.

ial
You should regularly clear
Windows NT Log Files the SYS$LOG.ERR and
VOL$LOG.ERR files to keep
Windows NT log files include the system log, the security log and the application them from using too much
log. You can display the contents of these log files by using the Windows NT server disk space. You can
Event Viewer. Each event log entry contains a header, a description of the event view the error logs from a
workstation by using a text

er
(based on the event type), and (optionally) additional data.
editor or from the console by
Five main event types are recorded in the log files, as described in the following using EDIT.NLM. You can
table. also view or clear the error
logs with the NETADMIN and
FILER utilities.
Type Description

at
Information Infrequent significant events that describe
successful operations of major services. For
example, a database program that loads

y
successfully might log an Information event.

op
Warning Events that aren’t necessarily significant but might
eM
indicate possible future problems. For example, a
Warning event might be logged if disk space is
low.

tC
Error Significant problem, such as a loss of data or
function. For example, an Error event might be
logged if a service didn’t load when the NT server
was booted.
Success Audit Audited security access attempts that were
No
successful. For example, a user’s successful
c

attempt to log on to the system might be logged

as a Success Audit event.
Failure Audit Audited security access attempts that failed. For
en

example, if a user tried to access a network drive

Do

and failed, the attempt might be logged as a

Failure Audit event.

Each type of event has a corresponding icon associated with it, which is dis-
se

played before the Date field in the event header (as displayed in Event Viewer).
r

Other fields include Time, Source, Category, Event, User, and Computer, so you
can tell when an event occurred; the process that caused the event; the event cat-
fe

egory (used primarily in the security log); which user, if any, initiated the
ea

process; and which computer the event occurred on. The Event field contains the
Event ID number, which identifies the type of event. You can get explanations of
Event IDs from the Help file for Event Viewer, or you can go to the Technet Web
site (www.microsoft.com/technet) and search for the phrase ″Event ID″ and the
Pl
Re

number of the event you want to research. Product support representatives can
use Event IDs and Sources to troubleshoot system problems.

Lesson 17: Troubleshooting the Network 439

 • The system log is similar to the NetWare SYS$LOG.ERR file, in that it
records system messages and alerts; however, it focuses on three main types
of events: Information, Warning, and Error. For example, the failure of a
driver or other system component to load during startup is recorded in the
system log.
• The security log records security events. This log helps track changes to the

ial
security system and identifies any attempts to breach security. The types of
events logged in this file are Success Audit and Failure Audit. The security
log can contain valid and invalid logon attempts as well as events related to
resource use, such as creating, opening, or deleting files or other objects. By
default, security logging is turned off. To enable security logging, use User
Manager for Domains to set the Audit policy.

er
• The application log contains events logged by applications. Event types
logged in the application log include Information, Warning, and Error. For
example, a database program might record a file error in the application log.
Application developers decide which events to monitor. Application logs
often contain additional data generated by the application that was the source

at
of the event record. Because the data appears in hexadecimal format, you’ll
probably need the assistance of a support technician who’s familiar with the
source application to be able to interpret the data.

y
Any user can view system and application logs, but only system administrators

op
eM
can view security logs.

Resources Provided by the Manufacturer

tC
If you gathered manufacturer resources such as readme files, support telephone
numbers, support CDs, and URLs for support Web sites as part of your network
documentation, as discussed in Lesson 16, you’ll find that these resources can
No
help you solve many network problems.
c

• Readme files contain information that might not be included anywhere else.
Many readme files contain a section describing known problems with the
software version being released, so make sure you check these files first
en

when you have a problem with network or application software.

Do

• Telephone support provides you with direct access to technical specialists

who can help you solve problems. Most software vendors have a toll-free
support number that you can use, but in the case of NOS software, the sup-
port itself might not be free—you’ll either buy a support contract or pay by
se

the support incident.

r

• Support CDs such as the TechNet and Novell Support Connection CDs con-
tain thousands of documents that you can search through to find answers to
fe

troubleshooting questions. In addition, these CDs contain updated files,

ea

patches, and drivers that you might need to implement to solve network
problems. Support CDs are updated periodically. The CDs aren’t free—
you’ll have to purchase a subscription.
Pl
Re

• Support Web sites such as TechNet (www.microsoft.com/technet) and Novell

Support Connection (support.novell.com) provide access to thousands of
technical documents, updated files, patches, and drivers to assist you in your
troubleshooting efforts. You don’t have to pay to use the Web sites, and the
information on the sites is updated constantly.

440
TASK 17B-1:
Identifying troubleshooting tools
1. Users throughout the company are complaining that network response
times are too slow. Which of the following should you use to help deter-

ial
mine the cause, and why?
a. Protocol analyzer and crossover cables.
b. Server console screen and protocol analyzer.
c. Performance monitors and cable testers.

er
d. Log files and performance monitors.

2. What’s the primary use for a loopback?

at
3. What tool would you use to solve IP addressing problems?

y
op
4. When is the most likely time to use a tone generator and tone locator?
eM
tC
APPLY YOUR KNOWLEDGE 17-2 Suggested time:
In this activity, you’ll work through some sample test questions that deal with the 30 minutes
information presented in this lesson.
No
c

1. Load the Skill Assessment component of the Network+ Certification

CD-ROM. (Remember to accept the disclaimer.)
en

2. Start the Network+ test.

Do

3. Select all items for the Troubleshooting the Network topic, uncheck the Test
Time check box, and then click OK.

4. Work through the sample questions and score the test.

se
r

5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
fe

and answers; then use the Scores button to return to your test score window.
ea

6. Exit the test.

Pl
Re

Lesson 17: Troubleshooting the Network 441

 Suggested time:
APPLY YOUR KNOWLEDGE 17-3
30 minutes In this activity, you’ll work through some sample test questions that deal with the
information presented in this course.

1. Load the Skill Assessment component of the Network+ Certification

ial
CD-ROM. (Remember to accept the disclaimer.)

2. Start the Network+ test.

3. Click OK to start a sample test with 20 randomly selected items from all
categories.

er
4. Answer each question, and score the test.

5. If you want to review the questions and the correct answers, click Review
Items. Use the Explanation and Next buttons to work through the questions
and answers; then use the Scores button to return to your test score window.

at
6. Exit the test.

y
7. Exit the Network+ Certification CD-ROM, and shut down the workstation.

op
eM
Summary
tC
In this lesson, you’ve identified procedures and tools that can help you solve
network problems.
No
c

LESSON 17 REVIEW
en
Do

17A List the steps of the Network+ troubleshooting model.

se
r

17B What types of network problems can hardware troubleshooting tools

help you detect?
fe
ea
Pl
Re

442
 SOLUTIONS

ial
LESSON 1 ANSWERS
Task/Step Question Response

er
Task 1A-1 Which of the following cor- a. Four computers that share a printer through
Question 1 rectly identifies a computer a printer-sharing device.
network?
b. Two computers connected by a temporary
serial-port connection and using file-transfer

at
software.
✓ c. A group of computers using hardware and

y
software to communicate and provide users
with access to shared resources.

op
eM d. Two computers, and their users, that share
information through the exchange of floppy
disks.

tC
Task 1A-1 A server is a computer that: ✓ a. Makes its resources available to other com-
Question 2 puters on the network.
b. Uses resources available from other comput-
ers on the network.
No
c

c. Uses resources from, and makes available its

resources to, other computers on the network.
en

Task 1A-1 What are the fundamental At least two computers connected by a communications
Do

Question 3 components of a network? link.

Task 1A-1 Compare the terms backbone Backbone is the communication channel that goes the
Question 4 and segment. entire length of the network and segments are communi-
cation channels connected to the backbone (or a portion
se

of the backbone).
r

Task 1A-1 In Figure 1-3, identify the Refer to Figure 1-2 for the answers.
fe

Question 5 segments and the backbone.

ea
Pl
Re

Solutions 443
LESSON 1 ANSWERS
Task/Step Question Response

ial
Task 1A-2 Identify factors that would Factors that discourage interconnectivity might include:
Question 1 make it difficult to intercon- • Differences in the type of communication channel used
nect two networks.
• Differences in the way that the communication channel
is accessed
• Differences in the way that data is encoded on the

er
communication channel
• Differences in the way that nodes connect to the
network
• Unreliability of data transfer between networks

at
Task 1A-2 Identify factors that would • Different network operating systems
Question 2 make it difficult for network
• Different protocols

y
computers to share files,
printers, and other resources. • Standards or the lack of standards

op
•
eM Differences in the way that data is presented in the
computer
• Differences in processors, which makes it difficult or
impossible for programs to run on different types of

tC
computers
• Differences in file and storage path-naming conventions

Task 1A-3 A. A large office building con- Because all computers are physically cabled together and
No
Question 1 tains 487 computers that are located in one geographic area, this is an example of a
c

cabled together and share LAN.

data, applications, and
hardware.
en
Do

B. A Utah-based branch office Because the computers aren’t located in one geographic
has five computers that share area, this is an example of a WAN, not a LAN.
data and a mail application
with two computers in an
office in Seattle.
se
r

Task 1A-3 A. You need to exchange files WAN. The purpose of a WAN is to connect all of a single
Question 2 and email with other employ- organization’s computers, regardless of the geographic
fe

ees in your company who are layout of the organization.

ea

located in offices throughout

the United States.
Pl

B. You need to share network LAN. LANs are relatively small computer networks con-
Re

applications and printers tained in a single building or campus.

with everyone in your depart-
ment; all of the desks are
within 100 feet of each other.

444
LESSON 1 ANSWERS
Task/Step Question Response

ial
C. You need to exchange files MAN. MANs are slightly larger than LANs and are con-
with other departments in tained within a single city or metropolitan area.
your company. All depart-
ments are in different
buildings throughout the city

er
and surrounding suburbs.

Task 1B-1 Covers a geographical area. Cellular

Question 1

at
Task 1B-1 Uses point-to-point connec- Mesh
Question 2 tion to every network device.

y
Task 1B-1 Workstations connect to the Ring
Question 3 network in a continuous loop.

op
Task 1B-1
eM
Uses drop cables to connect Bus
Question 4 to a backbone.

tC
Task 1B-1 Uses drop cables to connect Star
Question 5 to a central hub.

Task 1B-1 Uses multiple topologies. Hybrid

No
Question 6
c

Task 1B-1 Generates and sends signals Logical ring

Question 7 on a one-way path.
en

Task 1B-1 Generates and sends the sig- Logical bus

Do

Question 8 nal to all network devices.

Task 1B-1 The structure or layout of a Physical topology

Question 9 network’s transmission
media.
se
r

Task 1B-1 The path a signal follows Logical topology

fe

Question 10 through a network.

ea

Task 1C-1 You can take advantage of Mixed

Question 1 several different network
types.
Pl
Re

Task 1C-1 A mainframe computer ini- Hierarchical or legacy

Question 2 tiates and manages network
communication.

Solutions 445
LESSON 1 ANSWERS
Task/Step Question Response

ial
Task 1C-1 No single computer controls Peer-to-peer
Question 3 network communications.

Task 1C-1 A single computer provides Client-server

Question 4 specific network services to

er
clients.

Task 1C-1 You can access information Mixed

Question 5 on the Internet.

at
Topic Review Questions Response

What are the three compo- A network is made up of two or more nodes that share

y
A nents required to have a information, a communication medium through which
computer network? they can communicate, and network operating system

op
eM software.

Define physical topology and Physical topology defines the cable’s actual physical con-
B logical topology. figuration (star, bus, mesh, ring, cellular, hybrid). Logical

tC
topology defines the network path that a signal follows
(ring or bus).

What are the three relation- Hierarchical or legacy, client-server, and peer-to-peer.
No
C ships that exist between
c

network nodes?
en
Do

LESSON 2 ANSWERS
Task/Step Question Response
se
r

Task 2A-1 What are the three major Microsoft Windows NT, Novell’s NetWare, and UNIX
fe

Question 1 network operating systems? (and the various flavors including Linux).
ea

Task 2A-1 Who is the administrative The administrative users are Administrator for NT, Admin
Question 2 user for each of the three for NetWare 4 and above, SUPERVISOR for NetWare 3
major operating systems? and below, and root for UNIX systems.
Pl
Re

Task 2A-1 What are the core services The common services include file, print, security, messag-
Question 3 provided by the network ing, routing, and network administration. NT and
operating systems? NetWare also include directory services.

446
LESSON 2 ANSWERS
Task/Step Question Response

ial
Task 2B-1 Identify which systems in The Windows 95/98 or Windows NT Workstation systems
Question 1 your classroom are clients. are the clients on this network.

Task 2B-1 Your network includes Win- The systems come with the Client software for NT
Question 2 dows NT 4.0, NetWare 4.11, networks. NWClient 32 comes with NetWare and is then

er
and Linux 6.0 servers. What installed on the client machines. You need to configure
software would need to be TCP/IP to access the Linux server.
installed on your client in
order to communicate with
the network? Is that software

at
included with the client oper-
ating system or does it come
from the network operating

y
system?

op
Task 2B-1
Question 3
eM
What other clients could have
been included on this net-
Linux, DOS, Windows 3.x, and Macintosh clients could
also have been installed. They would all require software
work? What software would from Novell to communicate with the NetWare server; if
be required for the clients to the NT server is configured to run TCP/IP, the Linux and

tC
communicate with the class- Macintosh clients should be able to communicate with it.
room servers? The DOS and Windows 3.x clients would need additional
software in order to communicate with the NT server.

Task 2C-2 domain: A grouping of computers and users for administrative pur-
No
Question 1 poses
c

trust relationships: Establish communications between two or more domains

so that users can access resources in any domain.
en
Do

pass-through authentication: Enables a user to log on from a computer that is in a dif-

ferent domain. The user’s logon request is passed to the
appropriate trusted domain controller for validation.

Task 2C-2 With respect to management Answers might include:

se
r

Question 2 of user accounts, passwords,

and security, what are some • The ability to maintain only one account for each user
fe

of the goals you might have? that gives the user access to all network resources,
including Windows NT, NetWare, and SQL server;
ea

• The ability to synchronize the password for each user

account across all resources; and
• The centralized management of all network resources
Pl
Re

from one location.

Solutions 447
LESSON 2 ANSWERS
Task/Step Question Response

ial
Task 2C-2 How might you design the Answers might include: a single Windows NT domain
Question 3 domain structure and, if nec- that includes all servers and clients, or three domains, one
essary, the trust relationships at each geographic location. Trust relationships can be
for this network? established between each domain to give users access to
all network resources.

er
Check Your Skills How can you configure the You must establish a trust relationship between the San
2-1 network so that all users can Francisco and Los Angeles domains.
Question 1 log on from any computer at
either site?

at
Task 2C-3 What are some benefits to Answers might include: easy administration, simple con-
Question 1 implementing the single figuration (the default configuration when you install the

y
domain model? first domain controller), and centralized administration of
both users and network resources.

op
Task 2C-3
eM
How does administering a In the single master domain model, the administration of
Question 2 single master domain model user accounts is centralized in the master domain while
differ from administering a the administration of resources is decentralized in the

tC
single domain model? resource domains. In contrast, both user accounts and
resources are centrally administered in the single domain
model.
No
Task 2C-3 Why would you choose to You would choose the multiple master domain model if
Question 3 implement the multiple mas- your network requires more than 40,000 user accounts.
c

ter domain model instead of

the single master domain
model?
en
Do

Task 2C-3 How should you determine Because your network consists of more than one master
Question 4 where to create your users’ domain, you could create your users in any of the master
accounts in the multiple mas- domains. Some of the possible strategies you could use
ter domain model? for determining where to create users include basing your
design on the location of the users, or on the company
se
r

organization chart (usually by departments).

fe

Task 2C-3 When using the multiple The global groups must be created in each master domain
ea

Question 5 master domain model, where because they can contain accounts only from their own
should you create the neces- domain. You might have to duplicate global groups across
sary global groups? each of the master domains.
Pl
Re

Task 2C-3 Why would you choose to Answers might include: your company doesn’t have a
Question 6 implement the complete trust centralized MIS department and each domain is managed
domain model? by its own administrator; each department wants to main-
tain control over its own resources yet be able to grant
other domains’ users access to resources; or, your net-
work requires more than 40,000 user accounts.

448
LESSON 2 ANSWERS
Task/Step Question Response

ial
Task 2C-3 What is a risk of implement- Each domain must rely on other domains’ administrators
Question 7 ing the complete trust to configure their global groups with the correct users.
domain model?

Task 2C-4 What are some other ways The tree could also have been structured by shared

er
Question 1 the NDS tree in Figure 2-5 resources, by job levels, by plant location, or by floor.
might have been structured?

Task 2C-4 What is the main function of To provide users with access to network resources.
Question 2 the NDS tree?

at
Task 2C-4 What two things should you User access to network resources and ease of
Question 3 ensure that a tree design administration.

y
facilitates?

op
Task 2C-4
Question 4 nized?
eM
How can NDS trees be orga- By geographic location, organizational structure, adminis-
trative responsibilities, or any combination of these.

Task 2C-4 List some other things that These might include workflow, resource allocation, and

tC
Question 5 might affect a tree design. the flow of information.

Topic Review Questions Response

No
What are three popular net- Windows NT, NetWare, and UNIX.
c

A work operating systems?

en

What role does the client Users access the network using a client.
B
Do

have in a client-server net-

work?

What goals do directory ser- A single logon for a user to access all network resources
C vices try to attain? and the ability to centrally manage the network from a
single computer.
se
r

What is the Windows NT 4 It is called Windows NT Directory Services or NTDS and

fe

directory service called and it is implemented as a domain model.

ea

how is it implemented?

What is the Novell NetWare It is called Novell Directory Services and it is imple-
directory service called and mented as a hierarchical database, which is partitioned
Pl
Re

how is it implemented? and replicated throughout the network.

Solutions 449
LESSON 3 ANSWERS
Task/Step Question Response

ial
Task 3A-2 List the seven layers of the Application: User interface; user applications.
Question 1 OSI model and the function
Presentation: Data representation; data compression.
of each layer.
Session: Establish and maintain connections; deal with
upper layers.

er
Transport: Ensure reliability of delivered data; data
flow and transmission errors.
Network: Network addressing and routing; packet
assembly and disassembly.
Data Link: Mediate access to communication channel;

at
organize data in frames.
Physical: Specifications for physical channel; carry sig-
nals for all higher layers.

y
Task 3A-2 Create a mnemonic device to Examples might include: A popular song title needs darn-

op
Question 2
eM
help you remember the layers little publicity. Apply presently so that Ned doesn’t lose
(in order) of the OSI model. phaith.

Task 3B-1 Identify the family of proto- Remote-action.

tC
Question 1 cols that determines whether
a client or server completes a
process.
No
Task 3B-1 Identify the family of proto- Reliability.
c

Question 2 cols that ensures that data is

transferred without errors.
en

Task 3B-1 Identify the family of proto- Terminal emulation.

Do

Question 3 cols that enables a computer

to access a host node by
behaving like a terminal.

Task 3B-1 Identify the family of proto- Network management.

se

Question 4 cols that provides tools for

r

setting up and maintaining

the network.
fe
ea

Task 3B-1 Define the function of the Provide a way to ensure that data is transferred to the
Question 5 routing family of protocols. correct destination; determines the strategies used to
transmit data through the network.
Pl
Re

Task 3B-1 Define the function of the file Enable nodes to use network files; provides a common
Question 6 access family of protocols. means to access network files.

450
LESSON 3 ANSWERS
Task/Step Question Response

ial
Task 3B-1 Define the function of the Defines how data is represented; translates data for nodes
Question 7 codeset and data structure that use different coding schemes.
family of protocols.

Task 3B-1 Define the function of the Used to establish and maintain a connectionless or

er
Question 8 connection family of connection-oriented service for upper layers.
protocols.

Topic Review Questions Response

at
Describe the purpose of net- Architecture describes how components in the system
A work architecture and the behave and interrelate; the OSI model is a means for
OSI model. demonstrating architecture.

y
List the OSI layers and Application, Presentation, Session, Transport, Network,

op
describe how they
eM Data Link, and Physical. A layer requests services from
communicate. layers beneath it while providing services to the layers
above. Also, layers request services from and provide ser-
vices to corresponding layers in another stack.

B
Describe what is meant by
the families of protocols. tC
Groups of protocols that share common features.
No
c

LESSON 4 ANSWERS
en
Do

Task/Step Question Response

Task 4A-1 Can’t be tapped and result in Fiber optic cable.

se
r

Question 1 stolen data.

fe

Task 4A-1 A flexible cable that is easily PVC coaxial cable.

Question 2 installed, but shouldn’t be
ea

placed in crawl spaces.

Task 4A-1 Better for very high-speed, Fiber optic cable.

Pl
Re

Question 3 high-capacity data transmis-

sion because of the lack of
attenuation and the purity of
the signal.

Solutions 451
LESSON 4 ANSWERS
Task/Step Question Response

ial
Task 4A-1 The most commonly-used UTP twisted-pair cable.
Question 4 cable for existing telephone
systems; already installed in
many office buildings.

er
Task 4A-1 Uses a foil wrap between and STP twisted-pair cable.
Question 5 around wire pairs; uses inter-
nal twisting of the pairs to
give improved protection
from EMI.

at
Task 4B-1 Transmissions can’t occur Narrow-band radio, laser, and microwave
Question 1 through steel or load-bearing

y
walls.

op
Task 4B-1
Question 2
eM
Resistant to eavesdropping
and capable of high transmis-
Laser

sion rates; however, it is

susceptible to attenuation and

tC
interference.

Task 4B-1 Frequencies are divided into Spread-spectrum radio

Question 3 channels or hops.
No
Task 4B-1 Has a transmission delay of Satellite microwave
c

Question 4 typically 240 to 300

milliseconds.
en

Task 4B-1 Commonly used for long- Terrestrial microwave

Do

Question 5 distance voice and video

transmissions, and for short-
distance high-speed links
between buildings.
se
r

Task 4B-2 Your offices are located in an Spread-spectrum radio. Spread-spectrum radio can be
Question 1 historic 19th-century used where the installation of cabling is prohibited. It is
fe

building. You don’t want able to pass through walls.

ea

unsightly cables running

across the floor, but you are
not allowed to run cables
through the walls, floors, or
Pl
Re

ceiling.

452
LESSON 4 ANSWERS
Task/Step Question Response

ial
Task 4B-2 You need to provide a high- Point-to-point laser. Point-to-point laser is commonly used
Question 2 speed connection between two for short distance links between buildings. They are
offices that are separated by a capable of very high data rates, and are resistant to
courtyard. The offices are in eavesdropping.
separate buildings.

er
Task 4B-2 You need a high speed con- Fiber optic cable. Fiber optic cable has the highest trans-
Question 3 nection between graphic mission rate of any media, although it is the most
workstations that are in the expensive and complex bounded media system to install
same office. You have no and set up.

at
restrictions except the need
for speed.

y
Task 4B-2 You need to run a connection Fiber optic cable. Fiber optic cable is secure, and isn’t
Question 4 through a tunnel connecting affected by EMI.

op
eM
two buildings. The tunnel
contains many high voltage
lines and large electric
motors. You are concerned

tC
about security.

Task 4C-1
Question 1
No
The number Thousands (103) Hundreds (102) Tens (101) Ones (100)
c

4328 = 4 3 2 8
9488 =
en

101 =
Do

3=
90 =

9488: Thousands – 9; Hun-

se
r

dreds – 4; Tens – 8; Ones – 8.

101: Thousands – none; Hun-
dreds – 1; Tens – 0; Ones – 1.
fe

3: Thousands – none; Hun-

ea

dreds – none; Tens – none;

Ones – 3. 90: Thousands –
none; Hundreds – none; Tens
Pl

– 9; Ones – 0.
Re

Task 4C-1
Question 2

Solutions 453
LESSON 4 ANSWERS
Task/Step Question Response

ial
Decimal number Eights (23) Fours (22) Twos (21) Ones (20)
14 = 1 1 1 0
2=
12 =

er
7=
4=

2: Eights – none; Fours –

at
none; Twos – 1; Ones – 0. 12:
Eights – 1; Fours – 1; Twos –
0; Ones – 0. 7: Eights – none;

y
Fours – 1; Twos – 1; Ones –
1. 4: Eights – none; Fours –

op
1; Twos – 0; Ones – 0.
eM
Task 4C-1 How would you write the You would have to add an additional column to the table,
Question 3 decimal number 30 using the so as to create the binary number 11110.

tC
columns of the binary sys-
tem?

Task 4C-1 How many places (columns) You would need eight places, one each for the 128s, 64s,
No
Question 4 would you need to write the 32s, 16s, 8s, 4s, 2s and 1 places.
c

decimal number 255 as a

binary number?
en
Do
se
r
fe
ea
Pl
Re

454
LESSON 4 ANSWERS
Task/Step Question Response

ial
Task 4C-2 Identify whether the follow- The first, third, and fourth devices are analog; the second
Question 1 ing devices have analog or device is digital.
digital displays of informa-
tion:

er
at
y
op
eM
tC
No
c
en
Do
se
r
fe
ea
Pl
Re

Figure 4-15: Digital and analog devices.

Task 4C-2 List at least one analog Answers may include: blood pressure meter or standard
Question 2 device that is not displayed bathroom scale.
above.

Solutions 455
LESSON 4 ANSWERS
Task/Step Question Response

ial
Task 4C-2 List at least one digital device Answers might include: microwave oven display or VCR
Question 3 that is not displayed above. display.

Task 4D-1 Baseband Baseband: Uses a digital encoding technique to carry

Question 1 digital data over a digital transmission line; affected sig-

er
nificantly by attenuation; provides one communication
channel, over which only one computer can successfully
transmit at a time.

Task 4D-1 Broadband Broadband: Uses modulation techniques to transmit digi-

at
Question 2 tal data over analog carrier waves; less susceptible to
attenuation than baseband systems because of analog
signaling. Broadband media are uni-directional. There

y
must be two data paths.

op
Topic Review Questions
eM Response

Define bounded transmission A bounded transmission media is a wire or other physical

A media and give an example. conduit for an electrical or optical signal. Coaxial wire is

tC
one example of a bounded transmission medium.

Define unbounded transmis- An unbounded transmission media has no physical con-

B sion media and give an duit; rather, it uses the open air or electromagnetic
No
example. spectrum for signal transmission. Radio waves are one
c

type of unbounded transmission medium.

What are the two general Analog signaling: analog devices use a continuously vari-
C
en

approaches for representing able signal to represent portions of the data.

Do

data? Define each.

Digital signaling: digital devices use discrete signal levels
to represent portions of the data.

Identify and define the two Baseband and broadband transmissions. A baseband trans-
D techniques that can be used mission line provides only one channel per line, whereas
se
r

to transmit signals over the a broadband transmission line can provide multiple com-
data transmission media. munication channels.
fe
ea
Pl
Re

456
LESSON 5 ANSWERS
Task/Step Question Response

ial
Task 5A-1 Compare the operation of the The Physical layer operates on signals and the Data Link
Question 1 Physical layer and the Data layer operates on data.
Link layer.

Task 5A-1 Which networking and con- Bridges, switches, and NICs.

er
Question 2 nection devices work at the
Data Link layer?

Task 5A-1 What are the sublayers of the The Media Access Control (MAC) and Logical Link Con-
Question 3 Data Link layer? trol (LLC) sublayers.

at
Task 5B-1 What is the purpose of the To regulate the timing of data signals and eliminate
Question 1 MAC sub-layer? collisions.

y
Task 5B-1 What is a collision? The loss of data due to signals mixing together and

op
Question 2
eM becoming damaged beyond effectiveness.

Task 5B-1 Describe MAC addresses. The MAC address specifies a device’s actual physical
Question 3 address and is usually assigned by the hardware

tC
manufacturer. Every device on the network requires a
unique MAC address, which the MAC layer places on
frames sent out on the network.

Task 5B-2 What is the difference Synchronization at the Data Link layer accommodates
No
Question 1 between synchronization at bytes, characters, frames, and octets. Synchronization at
c

the Data Link layer and syn- the Physical layer accommodates only bits.
chronization at the Physical
layer?
en
Do

Task 5B-2 In asynchronous frame trans- It doesn’t occur. Transmitting and receiving clocks use
Question 2 missions, does comparable timing, but they’re never actually
synchronization occur? What synchronized. An example is typing on a computer
is an example of an asynchro- keyboard.
nous frame transmission?
se
r

Task 5B-2 If the parity bit is set to ″1,″ This would be even parity. The result would be 6.
fe

Question 3 what parity error detection is

being used in the example
ea

11010110? What is the result?

Task 5B-2 In the previous example, This would be odd parity. The result would be 5.
Pl
Re

Question 4 what would the result be if

the parity were set to ″0″?

Solutions 457
LESSON 5 ANSWERS
Task/Step Question Response

ial
Task 5B-2 In a CRC detection scheme, An algorithm, performed by both the transmitter and the
Question 5 what is used to determine receiver. Identical results indicate a good transmission.
whether a transmission is
good?

er
Check Your Skills Relies on only one network Isochronous
5-1 device to synchronize all net-
Question 1 work clocks.

Check Your Skills Sends a clock signal with Synchronous

at
5-1 every bit or character group.
Question 2

y
Check Your Skills Sends a clock signal with Asynchronous
5-1 every frame.

op
Question 3
eM
Task 5B-3 Describe the three types of Unacknowledged connectionless: LLC doesn’t acknowl-
Question 1 connections provided by edge that data has been successfully transferred. Upper

tC
LLC. layers must provide their own reliability checks. For this
reason, this type is very fast.
Connection oriented: This service establishes a dedicated
connection, enabling blocks of data to be transferred
No
between nodes until one of the nodes terminates the
c

connection.
Acknowledged connectionless: Similar to unacknowl-
en

edged connectionless, but it provides acknowledgement of

Do

individual frames.

Task 5B-3 Identify how a network that Upper-layer services that use this type of service must
Question 2 uses unacknowledged have some way to check the validity of data and to
connectionless service might request retransmission if the data isn’t valid.
cope with transmission
se
r

errors.
fe

Task 5B-4 Describe the use of sliding- Each frame isn’t acknowledged individually; instead, sev-
ea

Question 1 window flow control. eral frames are sent as a stream, and when the receiver’s
buffer fills, a choke packet is sent back telling the trans-
mitter to slow down. The transmitter does so, but then
slowly increases the stream again until another choke
Pl
Re

packet is received.

458
LESSON 5 ANSWERS
Task/Step Question Response

ial
Task 5B-4 Describe the use of static- A specific window size is determined and each frame in
Question 2 window flow control. the transmission is assigned a number up to the maximum
frame size. Then an acknowledgement is sent for each
frame, and no more frames can be sent until that same
number of acknowledgements has been received back.

er
Task 5B-4 Describe the use of stop-and- Each frame that is sent must be acknowledged before
Question 3 wait flow control. another frame can be sent.

Task 5B-4 List some instances that will Answers might include: frame fails to reach receiver,

at
Question 4 invoke LLC error control. offending frame size, differing checksums, capacity limit,
and interference.

y
Task 5C-1 You need to connect several 10Base-F, 100Base-T, or 10Broad36. These media types
Question 1 offices in a large industrial support long segments.

op
eM
complex that spans more
than half a mile.

Task 5C-1 You need to install a small 10Base2 or 10Base-T. Both are inexpensive and are

tC
Question 2 LAN in a single office. Your appropriate for small installations because they don’t sup-
budget limits how much you port high volume.
can spend.
No
Task 5C-1 You plan to install LAN 10Base2 isn’t as immune to EMF as optical fiber, but it
Question 3 cabling above the ceiling tiles would be better able to fit around tight curves than fiber.
c

in your office. There are

other electrical lines and fan
motors above the ceiling tiles,
en

and the cables must be run

Do

through several tight curves.

Task 5C-2 900 feet between segments; 2 Thin Ethernet.

Question 1 feet between T-connectors; 3
populated segments; and 25
se
r

connections per segment.

fe

Task 5C-2 300 feet between segments; Twisted-pair Ethernet.

ea

Question 2 350 total connections; star

topology; and 2 repeaters
between workstations.
Pl
Re

Task 5C-2 1,000 feet between segments; Thick Ethernet.

Question 3 2 populated segments; 100
devices per segment; and 2
repeaters between
workstations.

Solutions 459
LESSON 5 ANSWERS
Task/Step Question Response

ial
Task 5C-2 Which type of Ethernet is Thin. However, Thin Ethernet can be broken or damaged
Question 4 flexible? How flexible is it? from twisting or bending.

Check Your Skills How many repeaters can a 4

5-2 coaxial implementation have?

er
Question 1

Check Your Skills How many populated seg- 3

5-2 ments?
Question 2

at
Check Your Skills How many cable segments 5
5-2 per network?

y
Question 3

op
Task 5C-3
Question 1
Why was Fast Ethernet
developed?
eM To increase bandwidth to accommodate the advancements
of CPU performance, high-speed buses, and high-end
applications.

tC
Task 5C-3 What type of media does 100Base-TX: two pairs of Category 5 UTP or Category 1
Question 2 each of the Fast Ethernet STP cabling.
physical specifications
require? 100Base-T4: four pairs of Category 3, 4, or 5 UTP
cabling.
No
c

100Base-FX: two-stranded 62.5/125 micron multi-mode

or single-mode fiber.
en

Task 5C-3 True or false: A repeater that True.

Do

Question 3 translates frequencies

increases network latency.

Task 5C-4 What two things does IEEE That there be a token on the network at all times and that
Question 1 802.5 require? nodes pass it on when finished with it.
se
r

Task 5C-4 What happens if the token is Network transmission ceases because nodes can’t transmit
Question 2 not passed to the next node? unless they have the token.
fe
ea

Task 5C-4 What is the active monitor’s To constantly oversee the network to ensure that there is
Question 3 function? always a token being passed around.

Task 5C-4 True or false: It is not pos- False. You can set priority levels by using the priority
Pl
Re

Question 4 sible to prioritize node access control option.

to the network.

Task 5C-5 What is the most widely used Token ring.

Question 1 physical topology for IEEE
802.5?

460
LESSON 5 ANSWERS
Task/Step Question Response

ial
Task 5C-5 Why do 802.5 token ring Because relay switches within the MSAU skip over
Question 2 topologies look like star downed or unconnected nodes and the ring’s path
topologies? becomes obscured; the ring ″path” is hidden inside the
MSAU cabinet.

er
Task 5C-5 What is the purpose of a bea- To isolate the area in which a network error occurs.
Question 3 con frame?

Topic Review Questions Response

at
List some of the functions of Sharing the media; controlling data flow; packaging data
A the Data Link layer. in frames; correcting Physical-layer transmission errors.

y
What is the purpose of the Accessing the channel.

op
B MAC layer?
eM
Describe how communication LLC provides a standard interface, so the Network layer

tC
is handled with the upper doesn’t have to be concerned about lower-layer protocols
layers. that interface with LLC.

Describe LLC’s connection Unacknowledged connectionless service supports simple

No
services. point-to-point, multipoint, and broadcast transmission of
frames. This type of service is fast because LLC doesn’t
c

acknowledge that data has been successfully transferred.

Connection-oriented service establishes and maintains a
en

virtual connection between two stations.

Do

Acknowledged connectionless service combines features

of the other two services.

Provide a general description IEEE 802.3 supports network communication by using a

C of IEEE 802.3, and describe contention media-access method and a bus topology.
se
r

the differences between IEEE

802.3 and Ethernet. IEEE 802.3 spans the Physical and Data Link MAC lay-
fe

ers, whereas Ethernet spans from the Physical layer

ea

through the entire Data Link layer.

List the specifications of IEEE 802.5 supports network communication by using a

IEEE 802.5, and describe the token-passing media-access method and any topology,
Pl
Re

difference between IEEE although a star-wired ring is typically used.

802.5 and Token Ring.
IEEE 802.5 spans the Physical and Data Link MAC lay-
ers, whereas Token Ring spans from the Physical layer
through the entire Data Link layer.

Solutions 461
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6A-1 Converts digital signals to Modem
Question 1 analog and vice versa.

Task 6A-1 Runs several signals on one Multiplexer

Question 2 media segment.

er
Task 6A-1 Connects a device to a media NIC
Question 3 segment.

Task 6A-1 Receives signals from a Hub

Question 4 device and transmits them to

at
other devices.

Task 6A-1 Connects segments from dif- Bridge

y
Question 5 ferent networks.

op
Task 6A-1
eM
Connects segments from the Repeater
Question 6 same network and boosts
data signals.

Check Your Skills

6-1
Question 1
What is a media segment?

tC
A media segment is a section of transmission media that
has its own network address and enables network servers
and clients to access resources on a network.
No
Check Your Skills How does a passive hub dif- Answers might include: a passive hub doesn’t boost a
c

6-1 fer from an active hub? signal and different or additional restrictions can be
Question 2 placed on network segments attached to it.
en

Check Your Skills How does a switch differ A switch retransmits a signal by using only the specific
Do

6-1 from a multiport repeater? port through which the intended receiver can be reached.
Question 3 A multiport repeater retransmits a signal by using every
port.

Check Your Skills What is a NIC and why is it A NIC is a circuit board that can contain one or more
se
r

6-1 used? media connector types. It changes a computer’s signals to

Question 4 that which is compatible with the transmission medium. It
fe

is used to add a network interface board or port to a com-

puter that doesn’t have one built in.
ea

Apply Your McCloskey Associates needs Bridge. It enables filtering of unnecessary traffic, which
Knowledge 6-1 to provide occasional commu- largely comes from other segments.
Pl
Re

Question 1 nication between segments.

Although network traffic is
high, most communication is
among nodes on the same
segment.

462
LESSON 6 ANSWERS
Task/Step Question Response

ial
Apply Your Fenster Co., whose offices Repeater. Because traffic is as much across segments as it
Knowledge 6-1 occupy two floors of a small is within segments, a bridge would probably not add
Question 2 building, needs to extend the much benefit. (Switches, however, might enable traffic to
length of its Ethernet cable be configured better.)
system to connect two offices.

er
There are no standard net-
work traffic patterns.

Apply Your Condello Concrete Company Switch. Switches could be used to configure traffic flow
Knowledge 6-1 has an old Ethernet network. and facilitate configuration management.

at
Question 3 They would like to use their
existing network bandwidth
better and make it easier to

y
modify traffic patterns as
needed.

op
Apply Your
eM
KK Software Corp. has a Hub. Hubs act as a single connection point for several
Knowledge 6-1 room in which new software network devices.
Question 4 is tested. The eight computers

tC
in the room run their own
network and need a device
through which they can con-
nect to each other.
No
Task 6B-1 Interrupts Signals sent to the processor to gain its attention.
c

Question 1

Task 6B-1 I/O address A range of memory used to transfer data between an
en

Question 2 adapter and the CPU.

Do

Task 6B-1 Base memory address The starting address of memory on an adapter card.
Question 3

Task 6B-1 DMA channel The address of a communication channel between the
se
r

Question 4 adapter and the DMA co-processor.

fe

Task 6B-1 Driver A software program that enables a computer to communi-

ea

Question 5 cate with a device.

Pl
Re

Solutions 463
LESSON 6 ANSWERS
Task/Step Question Response

ial
Apply Your You just installed a new NIC Check to make sure that the card hasn’t been configured
Knowledge 6-3 in a workstation. This work- to use IRQ 3 or 4 because COM2 uses IRQ3 and COM 1
Question 1 station has a modem on and 3 use IRQ4. Since the workstation has no local print-
COM1, a mouse on COM2, ers attached, you could use IRQ5 or IRQ7.
and a plotter on COM3.

er
There are no local printers
on the workstation. The cus-
tomer has been unsuccessful
in connecting to the network.

at
Apply Your A network user got a second Check that the NIC isn’t using IRQ5 that is for LPT2 and
Knowledge 6-3 printer and now can no if it is, change to another IRQ.
Question 2 longer access the network.

y
Apply Your You just added a new system Check that the card is configured to use the appropriate

op
Knowledge 6-3
Question 3
eM
to the network that has a
3C509 combo card installed
media connector.

and you can’t access the

network.

Apply Your
Knowledge 6-3
Question 4
Why should you remove driv-
ers for devices no longer part
of your system? tC
To help ensure that there won’t be conflicts between old
drivers and any new drivers you install at a later time.
No
c
en
Do
se
r
fe
ea
Pl
Re

464
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6C-1 COMPUTER 01: This is the contention media access method, CSMA/CD.
Question 1 I’m ready to Drawbacks might include: it doesn’t guarantee computers
send. I see the access to the data-transmission media; a computer could
data transmission hog the channel, blocking the access of other computers;
media is free, the flow of data in heavy traffic increases collisions

er
so… because of additional computers on the network; devices
COMPUTER 01: can’t be assigned priorities for faster access.
Hi, COMPUTER
02! How are you
today? I’m done

at
sending…
COMPUTERS 02
and 03: I’m ready

y
to send. Whoops!
Someone else is

op
eM
sending, so I’ll
wait.
[Pause]

tC
COMPUTERS 02
and 03: I’m ready
to send. Whoops!
Someone else is
No
sending, so I’ll
wait.
c

[Pause]
COMPUTER 02:
en

I’m ready to
Do

send. I see the

data transmission
media is free, so...
COMPUTER 02:
Hi, COMPUTER
se
r

01! I’m doing

fine, thanks. I’m
done sending.
fe
ea
Pl
Re

Solutions 465
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6C-1 MEDIATOR: This is the polling media-access method.
Question 2 COMPUTER 01, Drawbacks might include: it ties up excessive bandwidth
do you have any- listening for and acknowledging messages.
thing to say?
COMPUTER 01:

er
No.
MEDIATOR:
COMPUTER 02,
do you have any-
thing to say?

at
COMPUTER 02:
Yes, as a matter
of fact. Fourscore

y
and...

op
MEDIATOR:
eM
Your time is up.
You’ll have to
finish later. COM-

tC
PUTER 01, do
you have anything
to say?
COMPUTER 01:
No
No.
c

MEDIATOR:
COMPUTER 02,
do you have any-
en

thing to say?
Do

COMPUTER 02:
Yes. (Where was
I?)...seven years
ago, our...
MEDIATOR:
se
r

Time’s up. COM-

PUTER 01, do
fe

you have anything

ea

to say?
COMPUTER 01:
No.
Pl
Re

466
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6C-1 COMPUTER 01: This is the token-passing media-access method.
Question 3 I’ve got the token. Drawbacks might include: inefficient use of the data
(Ahem!) Four- transmission media during low traffic; all devices use
score and ... complex software requiring intelligence.
Whoops! My

er
time’s up. Here’s
the token, COM-
PUTER 02.
COMPUTER 02:
Thanks, but I

at
don’t need it.
Here’s the token,
COMPUTER 03.

y
COMPUTER 03:
Thanks, but I

op
eM
don’t need it
either. Here’s the
token, COM-
PUTER 01.
COMPUTER 01:
Thanks. I’ve got
it. (Where was
I?)...seven years
tC
No
ago, our...
c

Whoops! Where
does the time go?
Here’s the token
en

COMPUTER 02.
Do

COMPUTER 02:
Thanks, but I still
don’t need it.
Here’s the token,
COMPUTER 03.
se
r

COMPUTER 03:
Thanks, but I
fe

don’t need it
ea

either. Here’s the

token, COM-
PUTER 01.
Pl
Re

Solutions 467
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6D-1 Ring topology.
Question 1

er
at
Task 6D-1 Bus topology.

y
Question 2

op
eM
tC
No
Task 6D-1 Hybrid topology.
c

Question 3
en
Do

Task 6D-1 Star topology.

se
r

Question 4
fe
ea
Pl
Re

468
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6D-2 Installation typically requires Bus topology.
Question 1 less data-transmission media
(cable) than all other
topologies.

er
Task 6D-2 All data goes through a cen- Star topology.
Question 2 tral point, where it is
distributed to the appropriate
address.

at
Task 6D-2 Uses one long cable to which Bus topology.
Question 3 network devices are either
directly attached or attached

y
by using short drop cables.

op
Task 6D-2
Question 4
eM
Several computers or devices
are joined together to form a
Ring topology.

circle.

tC
Task 6D-2 Messages move from one Ring topology.
Question 5 computer or device to the
next only in one direction.
No
Task 6D-2 Messages pass through all Ring topology.
Question 6 other computers on the way
c

to their destinations.

Task 6D-2 Uses a central hub with drop Star topology.

en

Question 7 cables extending in all

Do

directions.

Task 6E-1 Explain the 80/20 rule. It’s an efficiency guideline that suggests making 80% of
Question 1 your network traffic local and 20% of it internetwork,
because keeping traffic localized enables bridged seg-
se
r

ments to be used at the same time.

fe

Task 6E-1 How do a transparent They don’t. A transparent bridge performs typical bridge
ea

Question 2 bridge’s functions differ from functions, except that the transmitting device isn’t aware
those of a typical network that its data must travel through the bridge to get to its
bridge? destination.
Pl
Re

Task 6E-1 What service filters and for- A relay entity.

Question 3 wards data for a transparent
bridge?

Solutions 469
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6E-1 You have two redundant The data got caught in a bridging loop and couldn’t be
Question 4 bridges on your LAN for delivered to its destination. A bridging loop traps a data
fault tolerance. Data is sent packet in a continuous circular exchange between two
from one node to another but bridges because there are redundant paths within the
never arrives at its bridged network.

er
destination. What might have
happened? Why?

Check Your Skills Standby mode Blocking

6-2

at
Question 1

Check Your Skills Normal state Forwarding

y
6-2
Question 2

op
Check Your Skills Offline
eM Disabled
6-2
Question 3

Check Your Skills

6-2
Question 4
Unsettled network Listening

tC
No
Check Your Skills Gathering, but no forwarding Learning
c

6-2
Question 5
en

Task 6E-2 How does the spanning tree Its hierarchy—consisting of root, designated, and backup
Do

Question 1 protocol help prevent bridg- bridges—specifies one redundant bridge the designated
ing loops? bridge and the other backup, enabling fault tolerance.
Also, the redundant bridges can talk to each other about
route failures and dynamically re-route data.
se
r

Task 6E-2 During root bridge selection, The one with the lowest bridge ID.
Question 2 which network bridge
fe

becomes the root bridge?

ea

Task 6E-2 During designated bridge The one with the lowest port cost value.
Question 3 selection, which bridge
becomes the designated
Pl
Re

bridge?

Task 6E-2 When configuring the span- Yes.

Question 4 ning tree protocol, do you
want to have redundant
routes to each segment?

470
LESSON 6 ANSWERS
Task/Step Question Response

ial
Task 6E-2 What is the minimum num- Two.
Question 5 ber of bridges frames should
travel through?

Task 6E-2 What are some benefits of the Answers might include: fast topology changes, short

er
Question 6 spanning tree protocol? reconfiguration times for topology changes, and very little
latency.

Task 6E-3 What is the main difference Source-routing bridges use dynamic routing tables to
Question 1 between source-routing and route packets; transparent bridges use filtering databases.

at
transparent bridges?

Task 6E-3 Can route tables be stored in Yes.

y
Question 2 RAM?

op
Task 6E-3
Question 3 hello packet?
eM
What is the purpose of a To establish communication between two devices.

Topic Review Questions Response

A
Explain the purpose of net-
work connection devices and
describe issues associated
tC
Network connection devices connect networks and net-
work segments. As such, they are like traffic controllers:
they prepare packets to be transferred from one part of
No
with them. the network to another; and they determine which packets
c

should be forwarded and where they should be forwarded.

What is the role of a network Network adapters prepare data for the network cable,
B
en

adapter? send data to and receive data from the network, and
Do

translate data into bytes that the computer can understand.

What function does an access Access methods assure that only one computer at a time
C method provide for data can transmit data. They keep the sending and receiving of
transmissions? data over the network organized.
se
r
fe
ea
Pl
Re

Solutions 471
LESSON 6 ANSWERS
Topic Review Questions Response

ial
Draw an illustration of each
D of the following topologies:
Bus

er
at
y
op
eM
Star

tC
No
c
en
Do
se
r
fe
ea
Pl
Re

472
LESSON 6 ANSWERS
Topic Review Questions Response

ial
Ring

er
at
y
op
eM
E
How does a transparent
bridge forward data packets?
tC
If the bridge is in the Forwarding state, the bridge’s relay
entity checks the filtering database to determine whether
No
the packet should be forwarded or dropped.
c

What is a bridging loop and A bridging loop is the continuous cycling of a data packet
how does it occur? between two redundant bridges in which the packet is
en

never forwarded to its destination. It’s caused by having

Do

redundant paths in a transparent bridge environment.

How do source-routing Source-routing bridges use dynamic routing tables instead

bridges differ from transpar- of filtering databases.
ent bridges?
se
r
fe

LESSON 7 ANSWERS
ea
Pl
Re

Task/Step Question Response

Task 7A-1 Address of the cable only. Logical address.

Question 1

Solutions 473
LESSON 7 ANSWERS
Task/Step Question Response

ial
Task 7A-1 Combination of service, logi- Full address of entity.
Question 2 cal, device.

Task 7A-1 Address of the port or socket. Service address.

Question 3

er
Task 7A-1 Device processes. Entity.
Question 4

Task 7A-1 Address of the device only. MAC address.

at
Question 5

Check Your Skills Based on the structure of Network layer addresses and the associated processing is

y
7-1 addresses, explain why more complex than Data Link layer addresses.
Question 1 bridges are typically faster

op
than routers.
eM
Task 7B-1 Stone Investments needs to Brouter—enables internetwork communication in a
Question 1 provide communication mixed-protocol environment. A multi-protocol router

tC
between different networks. would also work.
Most of the networks use the
same protocols; however,
some use different ones.
No
Task 7B-1 Global Travel needs to con- Router—enable routers to determine the most efficient
c

Question 2 nect several networks that path for data across networks.
use the same upper-layer
protocols. They want to use
en

devices that will provide the

Do

most efficient path among the

networks.

Task 7B-1 Andy’s Supply House needs a CSU/DSU—enables public use of transmission media by
Question 3 high-speed WAN connection means of communication equipment that prepares signals
se
r

but can’t afford the high cost for transmission over a WAN.
of installation and mainte-
fe

nance that is associated with

ea

owning the communication

lines.

Task 7C-1 How are routes selected? The router can either randomly select a route (with pos-
Pl

sible priority ″weighting″) or sequentially and

Re

Question 1
systematically cycle through each one.

474
LESSON 7 ANSWERS
Task/Step Question Response

ial
Task 7C-1 What constitutes an A protocol that must route data by using static entries in
Question 2 unroutable protocol? List its lookup table; it is not capable of making its own rout-
some examples of unroutable ing decisions (typically LAN protocols). Examples
protocols. include NetBEUI, NetBIOS, LAT, and Routing SNA.

er
Task 7C-1 What type of connection A brouter.
Question 3 device should you use for
routing unroutable protocols?

Task 7C-2 What does static selection You manually enter the routes to be used into the route

at
Question 1 mean? tables.

Task 7C-2 What does dynamic selection Dynamic selection uses a router’s algorithms and calcula-

y
Question 2 mean? tions to perform the latest routing. The data stops at each
router in its path to determine the next best step.

op
Task 7C-2
eM
Compare static and dynamic Static routing is manually configured, and uses a pre-
Question 3 routing. scribed route that might or might not be the most efficient
route. Dynamic routing is automatically configured

tC
(except you must manually enter the first entry) and uses
route metrics to determine the most efficient route. Static
routing is more secure than dynamic routing since the
route is predictable.
No
Task 7C-3 Display the Gateway tab. Answers will vary. If there is no gateway listed, it’s
c

Question 3 What is the current gateway? because you are configured to use DHCP.
If there isn’t one listed, why
not?
en
Do

Task 7C-3 Display the IP Address tab. Is Answers will vary. If there is no IP address and subnet
Question 4 there a subnet mask? Why or mask listed, it’s because you are configured to use DHCP.
why not?

Topic Review Questions Response

se
r

Describe addressing associ- Entity—device processes; Logical address—address of the

fe

A ated with the Network layer cable; MAC address—device addresses; Internetwork
ea

of the OSI model. address—primary address of Network layer; Service

address—address of socket; Full address—combination of
service, logical, and physical addresses.
Pl
Re

Explain the purpose of net- Network connection devices connect networks and net-
B work connection devices and work segments. As such, they are like traffic controllers:
describe issues associated they prepare packets to be transferred from one part of
with them. the network to another; and they determine which packets
should be forwarded and where they should be forwarded.

Solutions 475
LESSON 7 ANSWERS
Topic Review Questions Response

ial
What are some of the NetBIOS, NetBEUI, Routing SNA, and LAT.
C unroutable protocols?

er
LESSON 8 ANSWERS
Task/Step Question Response

at
Task 8A-1 What are two other names Connection identifier and port.

y
Question 1 for ″socket″?

op
Task 8A-1
eM
A transaction consists of what A request and a response.
Question 2 two things?

Task 8A-1 List five service providers. File, message, print, database, and application servers.

tC
Question 3

Task 8A-1 What two addressing meth- The connection identifier and the transaction identifier.
Question 4 ods are in the Transport
No
layer?
c

Task 8A-1 Define connection ID in your A connection ID is an address that identifies (to upper
Question 5 own words. layers) each conversation being conducted by a particular
en

service.
Do

Task 8B-1 Segments are divided into False. Messages are divided into segments.
Question 1 messages.

Task 8B-1 Only one message can be sent False.

se

Question 2 out at a time.

r

Task 8B-1 A practical function of the True.

fe

Question 3 Transport Layer segment

ea

process is to help reduce

overhead.

Task 8B-1 Segment sequencing assigns True.

Pl
Re

Question 4 sequential numbers to seg-

ments before they are
transmitted. Upon arrival at
the destination, the out-of-
order segments are returned
to their original order.

476
LESSON 8 ANSWERS
Task/Step Question Response

ial
Task 8C-1 Define end-to-end flow An acknowledgment that permits the next communication.
Question 1 control.

Task 8C-1 List the four ways to control Make the duplicate number unique.
Question 2 errors.
Limit virtual circuits to a session.

er
Use only virtual circuits.
Drop packets that have been circulating too long through
a time-out process.

at
Task 8C-1 How do Go Back N and Both are error recovery processes in the Transport layer
Question 3 Selectively Repeat differ in that enable segments lost in transit to be requested for
strategy? re-transmission. Go Back N instructs the sender to

y
re-transmit an entire string from a specific packet while

op
eM Selectively Repeat requests re-transmission of only the
lost segments.

Task 8C-2 IP Connectionless

tC
Question 1

Task 8C-2 IPX Connectionless

Question 2
No
Task 8C-2 TCP Connection-oriented
c

Question 3

Task 8C-2 UDP Connectionless

en

Question 4
Do

Task 8C-2 SPX Connection-oriented

Question 5

Topic Review Questions Response

se
r

Describe the difference The connection identifier is an address to identify each

fe

A between the connection iden- line of conversation to the upper layers. The transaction
ea

tifier and transaction identifier is much like the connection identifier; however,
identifier addressing methods. it works with smaller pieces of information (transactions)
rather than a whole conversation.
Pl
Re

Describe the process of When a message is divided into segments in preparation

B segmenting. to depart on a packet-switched network, it undergoes a
synchronization process. During this process, a number is
placed on each segment for proper sequencing of all
pieces when they arrive at their destination.

Solutions 477
LESSON 8 ANSWERS
Topic Review Questions Response

ial
Define the three connection Unacknowledged connectionless service—doesn’t
C services in the Transport acknowledge data that has been successfully transferred.
layer.
Connection-oriented service—establishes and maintains a
virtual connection between two nodes.

er
Acknowledged connectionless service—combines features
of the other two services.

at
LESSON 9 ANSWERS

y
Task/Step Question Response

op
eM
Task 9A-1 Provide three examples of Typical examples include walkie-talkie, CB, and
Question 1 half-duplex transmission. intercom. (Telephone conversations generally occur in

tC
half-duplex, even though the telephone itself is capable of
full-duplex.) A debate is another example of a half-duplex
transmission, although if it deteriorates into a heated argu-
ment, it could become a full-duplex transmission.
No
Task 9A-1 Provide three examples of Some examples include broadcast radio, TV, and public-
c

Question 2 simplex transmission. address systems. A speech is another example of a

simplex transmission.
en

Task 9A-1 Provide three examples of Telephone, radar, smoke signals, and sign language are
Do

Question 3 full-duplex transmission. full-duplex transmissions.

Task 9A-1 What is the control mode Half-duplex.

Question 4 used by most LANs?
se
r

Task 9B-1 Acknowledgment of the Acknowledgement is part of step 2, transferring data.

Question 1 reception of data is part of
which managing step?
fe
ea

Task 9B-1 A connection release occurs It occurs when there is a loss of connection or when both
Question 2 in which two circumstances? parties agree to release the connection.
Pl
Re

478
LESSON 9 ANSWERS
Task/Step Question Response

ial
Task 9B-1 List the process for establish- Several criteria must be met before a connection can take
Question 3 ing a connection. place. First, login names must be verified with passwords.
Then, connection ID numbers are established, and the
duration is specified for the transmission. The party to
speak first is then negotiated. Finally, the Transport lay-

er
er’s acknowledgment numbering scheme and procedures
for re-transmission are determined.

Topic Review Questions Response

at
List and describe the three Simplex transmission enables only one device to transmit
A dialog control modes. at a time. All other devices function as receivers.

y
Full-duplex uses two communication channels so that
devices (usually two) on each end of a transmission can

op
eM transmit and receive data.
Half-duplex is a compromise between simplex and full-
duplex. In half-duplex mode, there is only one channel.

tC
List and briefly define the Establish the connection—involves negotiating the
B three steps for managing a communication.
session.
Transferring the data—maintains the connection.
No
Releasing the connection—ends that session.
c
en

LESSON 10 ANSWERS
Do

Task/Step Question Response

se
r

Task 10A-1 What does TCP/IP stand for? Transmission Control Protocol/Internet Protocol.
Question 1
fe
ea

Task 10A-1 What is a major advantage of It enables you to connect networks that contain all types
Question 2 using the TCP/IP protocol of computers.
suite?
Pl
Re

Task 10A-1 How does TCP/IP transmit By splitting the information into packets of data.
Question 3 data over networks?

Solutions 479
LESSON 10 ANSWERS
Task/Step Question Response

ial
Task 10A-1 What types of general ser- Answers might include: telnet, email, USEnet news, FTP
Question 4 vices are provided on the and anonymous FTP servers, WWW, and Gopher.
Internet?

Task 10A-2 Which DOD layer corre- The Host-to-Host layer.

er
Question 1 sponds to the OSI Transport
layer?

Task 10A-2 Which OSI layers correspond The Physical and Data Link layers.
Question 2 to the DOD Network Access

at
layer?

Task 10A-2 Which DOD layer is respon- The Internet layer.

y
Question 3 sible for routing packets
between computers or net-

op
works?
eM
Task 10A-2 Which DOD layer enables The Process/Application layer.
Question 4 users to run Internet applica-

tC
tions such as FTP?

Task 10B-1 What is DNS? The Domain Name System (DNS) is a shared database
Question 1 system that assigns or translates host names to IP
No
addresses for computers on an internetwork, and, more
specifically, the Internet.
c

Task 10B-1 What are some examples of Answers might include int, gov, mil, com, edu, net, and
Question 2 top-level domains? org.
en
Do

Task 10B-1 Create a fully qualified gns.com or rd.gns.com might be possible answers.
Question 3 domain name for the R&D
department of the GNS
corporation.
se
r

Task 10B-1 How can you get a domain If you use an ISP, get it through them; or, apply for a
Question 4 name? name through InterNIC.
fe
ea

Task 10B-2 What is the purpose of the It uniquely identifies each computer on a network
Question 1 host portion of the IP segment.
address?
Pl
Re

Task 10B-2 If your IP address is 175.135. The IP address should be 175.135.host.host, where the
Question 2 200.1, what must the IP host address portion can be anything but 200.1, 200.0, or
address of the network card 200.255.
in the router be?

480
LESSON 10 ANSWERS
Task/Step Question Response

ial
Task 10B-3 Is 221.100.155.0 a valid IP No. It is a Class C address, which means that only the
Question 1 address? Explain why. last byte identifies the host. A host address of 0 isn’t per-
mitted since the resulting IP address is actually the
network address for the segment.

er
Task 10B-3 Is 187.131.255.5 a valid IP Yes. The network address is 187.131 and the host address
Question 2 address? Explain why. is 255.5, both of which conform to the Class B IP
addressing rules.

Task 10B-3 Is 121.0.0.1 a valid IP Yes. The network address is 121 and the host address is

at
Question 3 address? Explain why. 0.0.1, both of which conform to the Class A IP addressing
rules.

y
Task 10B-4 What types of systems on Answers might include: workstations, file servers, UNIX
Question 1 your network require host hosts, printers with network cards, and routers.

op
addresses?
eM
Task 10B-4 What are some guidelines you Answers might include: designating one of the bytes of
Question 2 have implemented or are the host address to identify the type of operating system,

tC
thinking of implementing for or using a range of addresses for the last byte of the host
assigning host addresses on address to differentiate between workstations and routers.
your own network?
No
Task 10B-5 Compare using host tables Host tables are manually maintained ASCII lists copied to
Question 1 and DNS. each system on the network so that you can use names to
c

refer to systems and the host tables are used to look up

the IP address associated with the name. Every system
must have a current copy of the table and any changes
en

must be made to all copies of the file. Therefore, it is

Do

only useful on small networks without Internet or intranet

access. DNS is a service used to resolve domain names to
addresses on the Internet.

Task 10B-5 What is a DNS zone? A DNS zone is a section of the DNS database that con-
se
r

Question 2 tains domains. Each zone is managed by its organization.

fe

Task 10B-5 What is another name for a Resolver.

ea

Question 3 DNS client that submits

requests for IP addresses?

Task 10B-6 After viewing the presenta- DNS acts as an electronic address book for TCP/IP
Pl
Re

Question 3 tion, list at least two ideas addresses; the hierarchy of the DNS naming system was
presented in the video. presented; purpose of DNS was covered; used an email
example to explain how names are resolved on the
Internet.

Solutions 481
LESSON 10 ANSWERS
Task/Step Question Response

ial
Task 10B-8 What is a daemon? A server program.
Question 1

Task 10B-8 Which protocol provides Telnet.

Question 2 terminal-emulation function-

er
ality?

Task 10B-8 Which protocol provides SMTP.

Question 3 email functionality?

at
Task 10B-8 Which protocols enable you FTP and TFTP.
Question 4 to transfer files from host to
host?

y
Task 10B-8 What is the main difference TCP provides reliable connections; UDP doesn’t.

op
Question 5 between TCP and UDP?
eM
Task 10B-8 Name an application that Answers include Telnet, FTP, and SMTP.
Question 6 uses TCP.

Task 10B-8
Question 7
Name an application that
uses UDP.
tC
Answers include NFS and TFTP.
No
Task 10B-8 Which protocol deals with ICMP.
Question 8 error and control messages?
c

Task 10B-8 Which protocol is responsible IP.

Question 9 for packet fragmentation,
en

routing and assembly?

Do

Task 10B-8 Which two protocols are used RARP and BOOTP.
Question 10 by diskless workstations?

Task 10B-8 Which protocol translates ARP.

se
r

Question 11 software addresses to MAC

addresses?
fe
ea

Task 10B-8 What types of specifications Specifications for boards, connectors, cabling, and net-
Question 12 are defined at the Network work topology.
Access layer?
Pl
Re

Task 10B-8 Identify the following data Frames correspond to the Network Access layer, packets
Question 13 unit names with the appro- to the Internet layer, data to the Process/Application layer,
priate DOD layers: frame, and message to the Host-to-Host layer.
packet, data, and message.

482
LESSON 10 ANSWERS
Topic Review Questions Response

ial
Describe how TCP/IP com- At the transmitting workstation, TCP/IP breaks the infor-
A municates information. mation down into separate network packets, sends them
individually, and then reassembles them into the original
information at the target workstation.

er
Where did TCP/IP originate, TCP/IP began as an experiment by the Department of
and what was its purpose? Defense to interconnect diverse computer systems.

List several of the well- Answers might include: 21 FTP, 23 Telnet, 25 SMTP, 53
B known ports and the DNS, 69 TFTP, 79 finger, 80 HTTP, 110 POP3, 119
application related to that NNTP, 161 SNMP.

at
port.

y
op
LESSON 11 ANSWERS
eM
tC
Task/Step Question Response

Task 11A-1 01100100 100

Question 1
No
c

Task 11A-1 11100000 224

Question 2
en

Task 11A-1 11111111.11111111.11110000. 255.255.240.0

Do

Question 3 00000000

Task 11A-1 01100100.01100100.00000010. 100.100.2.1

Question 4 00000001
se

Task 11A-1 011111111.00000000.00000000. 127.0.0.1

r

Question 5 00000001
fe

Task 11A-2 255 11111111

ea

Question 1

Task 11A-2 8 00001000

Question 2
Pl
Re

Task 11A-2 198.131.205.5 11000110.100000011.11001101.00000101

Question 3

Solutions 483
LESSON 11 ANSWERS
Task/Step Question Response

ial
Task 11A-2 224.254.100.8 11100000.11111110.01100100.00001000
Question 4

Task 11A-3 If your network consisted of Class A: one network with many hosts.
Question 1 200,000 individual computers

er
worldwide, what class of IP
address would you need to
directly connect each com-
puter to the Internet?

An alias for ″this network,″ it is typically used for testing

at
Task 11A-3 What is the network address
Question 2 of 127 used for? the configuration of TCP/IP.

y
Task 11A-3 If you plan to connect your You must write to the InterNIC to obtain an IP address.
Question 3 network to the Internet, what You can’t choose your own network address. You must

op
ing your IP addresses?
eM
are the restrictions for assign- choose a unique host address for each machine on the
Internet.

Task 11A-3 How do you calculate the 2 raised to the number of bits available to vary. Because

tC
Question 4 available number of host Class B addresses have 2 bytes or 16 bits available for
addresses for a Class B the host address, the formula is 2 16 -2 or 65,534.
address?
No
Task 11A-3 Why is a host address of 255 The 255 address is used to identify broadcasts on the
Question 5 not allowed? network.
c

Task 11A-3 Why is a host address of 0 The 0 address is used in combination with the network
Question 6 not allowed? address to identify network segments for routing
en

purposes.
Do

Task 11B-1 Define autonomous system. An autonomous system is a group of routers that route
Question 1 packets by using the same IGP and are managed by one
authority.
se
r

Task 11B-1 Which IGP uses RIP as its Distance vector.

Question 2 routing protocol?
fe
ea

Task 11B-1 How does the split horizon By broadcasting its routes to all routers except the one
Question 3 method relieve the count-to- from which it learned its own location; inaccurate infor-
infinity problem? mation is not passed around the network.
Pl
Re

Task 11B-1 Link-state routers broadcast True.

Question 4 to only those devices to which
they are directly connected.

484
LESSON 11 ANSWERS
Task/Step Question Response

ial
Task 11B-2 What means does OSPF use Hello packets.
Question 1 to establish adjacencies
between routers on the same
segment?

er
Task 11B-2 How are DRs and BDRs They are elected according to priority and the router with
Question 2 elected and which router the highest priority becomes the DR.
becomes the DR?

Task 11B-2 What happens if you assign a It can’t become a DR or a BDR.

at
Question 3 router a priority of zero?

Task 11B-2 What is the difference During the two-way state, routers have exchanged data-

y
Question 4 between a two-way state and base information, but synchronization hasn’t yet taken
a full neighbor state? place. During the full-neighbor state, routers have syn-

op
eM chronized database information and established adjacency
to each other.

Task 11C-1 Why would you subnet your Because you have been given only one network address

tC
Question 1 network? by the InterNIC; you haven’t been given enough
addresses to connect all of your hosts to the Internet. Or,
because you have several physical nnetwork segments.
No
Task 11C-1 Why would you segment your Answers might include: to reduce the number of
Question 2 network? machines on a single segment in order to reduce network
c

traffic; to overcome physical limitations such as exceeding

the maximum number of machines on a given segment; to
connect multiple topologies, such as Ethernet and Token
en

Ring, together via routers.

Do

Task 11C-1 If your network consists of Three.

Question 3 three segments connected by
two routers, how many
subnet (network) addresses
se
r

will be necessary for proper

configuration of TCP/IP?
fe
ea

Task 11C-1 If your network consists of 62—one for each of the 30 hosts on each of the two seg-
Question 4 two segments, with 30 hosts ments and two for the network cards in the router.
each, connected by a router,
how many host addresses will
Pl
Re

you need?

Task 11C-2 Which portion of the IP The host address.

Question 1 address does the subnet mask
hide?

Solutions 485
LESSON 11 ANSWERS
Task/Step Question Response

ial
Task 11C-2 Why do you need a subnet The subnet mask is the only way TCP/IP can determine if
Question 2 mask if your company’s net- a packet belongs to the local or to a remote network.
work doesn’t connect to the Without a subnet mask, a network couldn’t be divided
Internet? into smaller subnets because TCP/IP wouldn’t be able to
route packets.

er
Task 11C-2 What other function does the The subnet mask enables you to borrow bits from the
Question 3 subnet mask perform in the host address bytes to further subdivide the IP network
event your network connects addresses assigned by the InterNIC.
to the Internet?

at
Task 11C-3 When is the default subnet You would use the default subnet mask if your network
Question 1 mask used? doesn’t connect to the Internet or if your network con-

y
nects to the Internet but consists of only one segment.

op
Task 11C-3
Question 2
eM
If your IP address is 110.105.
25.0, what is the default
The Class A default subnet mask, 255.0.0.0.

subnet mask?

tC
Task 11C-3 If your IP address is 198.105. The Class C default subnet mask, 255.255.255.0.
Question 3 25.0, what is the default
subnet mask?
No
Task 11C-3 If your IP address is 135.64.4. The Class B default subnet mask, 255.255.0.0.
Question 4 10, what is the default subnet
c

mask?

Task 11C-4 What is the purpose of a cus- To subdivide the InterNIC-assigned IP network address
en

Question 1 tom subnet mask? into separate subnet addresses.

Do

Task 11C-4 If you borrow three bits from Subnet addresses=2 3 -2=6.
Question 2 the host portion of a Class B
address, how many subnets
will you be able to assign
se
r

unique addresses to?

fe

Task 11C-4 How many host addresses Host addresses=2 13 -2=8,190.

ea

Question 3 will still be available after

subnetting?

Task 11C-4 If your IP address is 145.8. 145.8.17.96 is a Class B address; the subnet mask is bor-
Pl
Re

Question 4 17.96 and the subnet mask is rowing the entire third byte, so the subnet address is 17.
255.255.255.0, what is the
subnet address?

Task 11D-1 What is the purpose of The purpose of supernetting is to prevent the depletion of
Question 1 supernetting? network addresses.

486
LESSON 11 ANSWERS
Task/Step Question Response

ial
Task 11D-1 How does supernetting sup- By eliminating the traditional concepts of Class A, Class
Question 2 port the deployment of B, and Class C network addresses and replacing them
arbitrarily sized networks? with the generalized concept of the network prefix. Rout-
ers use the network prefix, rather than the first three bits
of the IP addresses, to determine the dividing point

er
between the network ID and the host ID, enabling net-
work sizes to vary.

Topic Review Questions Response

at
Why are IP addresses divided They are divided into classes so that the Internet can sup-
A into classes? port various network types, such as those with few
networks but many hosts, more networks but fewer hosts,

y
and many networks but minimal hosts. Portions of the IP
address represent a class, which identifies a device as

op
eM either a host or network.

Describe IP’s two routing Distance vector: LAN routers periodically broadcast their
B methods. entire route tables (containing distances from routers to

tC
other LANs). They use the Routing Internet Protocol
(RIP).
Link-state: LAN routers broadcast route information for
No
only those interfaces directly connected to them. They
c
typically use the Open Shortest Path First (OSPF)
protocol.

List some advantages of

en

• Reduces network traffic.

C subnetting a network.
Do

• Connects disparate network topologies.

• Increases network security.
• Minimizes troubleshooting.
• Decreases physical network limitations.
se

• Connects networks across geographical areas.

r

What is the main purpose of To disguise the host section of an IP address in order to
fe

a subnet mask? determine whether a packet’s destination is local or

ea

remote.

Describe a supernet. The supernet strategy uses a network prefix to specify the
D
Pl

dividing point between the network and host portions of

Re

the address, enabling networks of variable sizes.

Supernetting is a strategy to prevent the depletion of
Internet addresses.

Solutions 487
LESSON 12 ANSWERS
Task/Step Question Response

ial
Task 12A-2 Display the DNS tab. Is If Enable DNS is selected, DNS is being used to resolve
Question 3 Enable DNS selected? Why addresses to their IP address.
or why not?

Task 12A-2 Display the IP Address tab. Is It means that you are getting your IP address through

er
Question 5 Obtain An IP Address Auto- DHCP.
matically selected? If so, what
does this mean?

Task 12B-1 Enter ARP. It will display a -a or -g: Displays current ARP entries by interrogating the
Question 2 summary of the command current protocol data. If inet_addr is specified, the IP and

at
and list the options used with Physical addresses for only the specified computer are
the command. Record the displayed. If more than one network interface uses ARP,
options and a brief descrip- entries for each ARP table are displayed.

y
tion of each.
Inet_addr: Specifies an Internet address.

op
eM -N if_addr: Displays the ARP entries for the network
interface specified by if_addr.

tC
-d: Deletes the host specified by inet_addr.
-s: Adds the host and associates the Internet address inet_
addr with the physical address eth_addr. The physical
address is given as 6 hexadecimal bytes separated by
No
hyphens. The entry is permanent.
c

Eth_addr: Specifies a physical address.

If_addr: If present, this specifies the Internet address of
en

the interface whose address translation table should be

Do

modified.

Task 12B-1 Which options have the same -a and -g.

Question 4 function?
se

Task 12B-3 What are get, put, ls, open, get—receive a file.
r

Question 5 and bye used for?

put—send a file.
fe
ea

ls—list the contents of the remote directory.

open—connect to remote FTP site.
bye—end an FTP session and exit the FTP utility.
Pl
Re

488
LESSON 12 ANSWERS
Task/Step Question Response

ial
Task 12B-6 Record the available options -a—all connections and listening ports.
Question 3 and their uses.
-e—Ethernet statistics.
-n—addresses and ports in numeric form.

er
-p protocol—information on the specified protocol (TCP,
UDP, or IP).
-r—routing.
-s—per protocol statistics.

at
interval—how many seconds between updating
information.

y
Task 12B-7 What other information can Answers might include: subnet mask; adapter address;

op
Question 4 you find about your TCP/IP default gateway address; host name; DNS, DHCP, and
eM
configuration using the WINS information; and the ability to release or renew
appropriate utility? dynamically assigned IP addresses for Windows 95/98/NT
systems. For UNIX systems, link encap type, hardware

tC
address, broadcast address, subnet mask, information on
RX and TX packets, or changing the current
configuration.

Task 12C-1 You’re working the helpdesk From the problem workstation, enter ARP -a to list the
No
Question 1 and get a call that a user ARP cache.
c

can’t access the UNIX host at

150.150.32.157. You are on
the same subnet as the user
en

and the UNIX host and try to

Do

ping the UNIX host. You can

successfully do so. You can
also ping the user’s
workstation. When you ask
the user to enter ping 150.
se
r

150.32.157, all they get is a

series of Destination Unreach-
able messages. What should
fe

you do?
ea

Task 12C-1 The results of the listing dis- From the problem workstation, enter ARP -d 150.150.32.
Question 2 played from step 1 shows the 157 to delete this entry from the ARP cache, and then add
Pl

UNIX host, but the MAC the corrected entry using the command ARP -a 150.150.
Re

address listed isn’t the MAC 32.157 mac_addr.

address for the NIC in that
system. What should you do?

Solutions 489
LESSON 12 ANSWERS
Task/Step Question Response

ial
Task 12C-1 You’re a network administra- You could log in to the FTP server and enter NETSTAT
Question 3 tor and have been receiving -a to see if it is listening on Port 21. You might also try
complaints that users aren’t to telnet into the system to see if you get a ″ready″
able to post files to or down- message.
load files from the FTP

er
server. What might you do to
determine the status of the
server?

Task 12C-1 A Windows 95 user logs in Use the winipcfg command to display the current TCP/IP

at
Question 4 and gets a message that their configuration information, click More Info, and then click
IP address is already in use. the Release or Release All button. Then restart the work-
This user receives their IP station so they can receive a new IP address.

y
address through a DHCP
server. How can the user get

op
eM
a valid IP address so they can
do their work?

Task 12C-1 A client calls from the Cali- Try pinging the server from your workstation. If you are

tC
Question 5 fornia office saying that they able to successfully reach the server, ask the client to do
aren’t able to connect to the the same and see if they get a response that the server is
server ICANY in New York. available. (They’ll most likely get a Destination Host
This server is on a routed IP Unreachable message.) If they can’t connect, have the
network. This is the second client enter tracert icany to see how far they can reach
No
client from California who before receiving the Destination Unreachable message.
c

has called with this problem. You’ll then be able to figure out which device is causing
No users from other sites the problem, because the next device that would be
have called. What should you encountered on the route between the client and ICANY
en

do? is the problem device.

Do

Task 12C-1 You need to download the FTP.

Question 6 latest patches for several
operating systems and soft-
ware applications. Which
se
r

troubleshooting tool would be

best suited to this task?
fe
ea

Apply Your arp -a

Knowledge 12-1
Question 1
Pl
Re

Apply Your nbtstat -s

Knowledge 12-1
Question 2

490
LESSON 12 ANSWERS
Task/Step Question Response

ial
Apply Your tracert ziff.net
Knowledge 12-1
Question 3

er
Apply Your netstat

at
Knowledge 12-1
Question 4

y
Apply Your ipconfig /all
Knowledge 12-1

op
Question 5
eM
Apply Your
Knowledge 12-1
tC
winipcfg, then click on More Info, or winipcfg /all
No
Question 6
c
en
Do
se
r
fe
ea

Apply Your ping 155.40.105.199

Knowledge 12-1
Question 7
Pl
Re

Apply Your ftp lo.tool.net

Knowledge 12-1
Question 8

Solutions 491
LESSON 12 ANSWERS
Task/Step Question Response

ial
Apply Your telnet
Knowledge 12-1
Question 9

er
at
Apply Your ifconfig
Knowledge 12-1
Question 10

y
op
Topic Review Questions
eM Response

What do you need to config- The IP Address and subnet mask, DNS, the default gate-

tC
A ure on your workstation to way, any additional gateways used on internetworks,
establish a TCP/IP network WINS (if you’re using a WINS Service), and the com-
connection? puter name.
No
Identify the purpose of each ARP is used to display and modify the local ARP cache.
B
c

of the following TCP/IP suite

utilities: ARP, Telnet, Telnet is used to remotely connect to the server via termi-
NBSTAT, TRACERT, nal emulation for administration and troubleshooting
purposes.
en

NETSTAT, IPCONFIG,
Do

WINIPCFG, FTP, and PING. NBSTAT is used to check NetBIOS state over TCP/IP
connections.
TREACERT is used to check the route to a remote
system.
se
r

IPCONFG is used on Windows NT systems to display the

current IP configuration.
fe

WINIPCFG is used on Windows 95/98 systems to display

ea

the current IP configuration.

FTP is used to transfer files between systems.
Pl

PING is used to verify that two computers can

Re

communicate.

What utility would you use to TRACERT

C find a problem in the path
between two IP addresses?

492
LESSON 12 ANSWERS
Topic Review Questions Response

ial
What utilitity would you use PING
to make sure an IP address is
reachable?

er
LESSON 13 ANSWERS

at
Task/Step Question Response

Task 13A-1 Compare SLIP and PPP SLIP IP addresses must be configured each time a con-

y
Question 1 connections. nection is established. PPP IP addresses can be
dynamically configured. SLIP requires more modem script

op
eM configuration to establish the connection. PPP has less
overhead than SLIP to establish the connection. PPP can
be used for more secure transmissions than SLIP
connections.

Task 13A-1
Question 2
Which protocol would you
need to use if you are using a
VPN?
PPTP.
tC
No
c

Task 13A-1 Which protocol is capable of PPP. (Remember: PPTP is encapsulated within PPP and
Question 3 transmitting multiple proto- PPP actually does the transmission.)
cols over a serial link?
en
Do

Task 13A-1 When would you use PPTP? When you need to establish secure communication over
Question 4 standard Internet connections.

Task 13B-1 What equipment is needed You’ll need an NT1 device and possibly other ISDN
Question 1 for ISDN connections? interfaces and ISDN LAN topology bridges.
se
r

Task 13B-1 Compare European with BRI is the same. PRI provides up to 1,536 Kbps with the
Question 2 American and Japanese American and Japanese standard. The European standard
fe

ISDN connections. PRI provides up to 1,984 Kbps. Different interfaces are

ea

used for the two standards.

Task 13B-1 Compare the use of analog Analog lines require a modem to translate digital signals
Pl

Question 3 lines with ISDN lines. to analog and back to digital; ISDN lines are end-to-end
Re

digital transmissions without the need for translation.

Analog requires separate lines for simultaneous voice and
data transmission; ISDN can use a single line for this.

Solutions 493
LESSON 13 ANSWERS
Task/Step Question Response

ial
Task 13B-1 What are the functions of the The B channels carry data and the D channel controls the
Question 4 B and D channels used in transmission over the B channels.
ISDN?

Task 13B-1 How many channels are each T1 is composed of 24, T2 of 96, T3 of 672, and T4 of

er
Question 5 of the T-carriers composed 4,032 channels.
of?

Task 13B-1 What equipment is used with CSU/DSU equipment is used.

Question 6 Digital Data Service rather

at
than use of modems?

Task 13B-1 Compare ATM and SMDS They both use fixed-length packets called cells. ATM uses

y
Question 7 services. error-checking and SMDS doesn’t.

op
Topic Review Questions
eM Response

List the serial protocols dis- SLIP, PPP, PPTP.

A cussed in order from least

tC
secure to most secure.

Describe characteristics of POTS are analog connections requiring a modem. The

B POTS and ISDN connections. FCC limits their speed to 53kbps.
No
c

ISDN are digital connections requiring NT1 devices

rather than modems for connections. There are two 64
Kbps BRI channels and 23 64 Kbps PRI channels. B
channels carry data and the D channel controls the trans-
en
Do

mission over the B channels.

If Dial-up Networking isn’t From the Add/Remove Programs utility in Control Panel,
C installed on your workstation, select the Windows Setup tab, then select
how do you install it? Communications. Click Details, and then check the
Dial-Up Networking option.
se
r
fe
ea

LESSON 14 ANSWERS
Pl
Re

Task/Step Question Response

Task 14A-3 What are the levels of Login, file system, and NDS.
Question 1 NetWare security in a
NetWare 4.11 network?

494
LESSON 14 ANSWERS
Task/Step Question Response

ial
Task 14A-3 What command is used to set chmod.
Question 2 permissions on a UNIX sys-
tem?

Task 14A-3 What are the levels of Win- User and share.

er
Question 3 dows security?

Task 14B-1 What are passwords? A way to electronically protect data by requiring users to
Question 1 enter a string of characters to identify themselves to the
system.

at
Task 14B-1 How is a password policy Through the Windows NT Account Policy.
Question 2 implemented in an NT envi-

y
ronment?

op
Task 14B-1
Question 3
eM
What are some areas that a
password guidelines docu-
Issues might include minimum password length, fre-
quency of password changes, whether to allow users to
ment might cover? reuse passwords, how many attempts a user has to log in
without a valid password, and who sets the passwords.

Task 14B-1
Question 4
List some examples of good
passwords. List some
examples of poor passwords. tC
Good examples might include Trng_1, JPUT_3B99,
$5nd3irt4. If you’re using UNIX or another operating
system that recognizes case-sensitive passwords, mixing
No
upper- and lower-case letters in your password helps
make it more secure.
c

Poor examples might include password, fido, Gail,

1234567890.
en
Do

Task 14C-1 What is data encryption? The process of converting data into coded form in such a
Question 1 way that only authorized parties can access the
information.

Task 14C-1 Compare stream cipher and Stream cipher encrypts data a bit at a time and produces
se

Question 2 block cipher. fewer errors than other methods. Block cipher encrypts
r

data a block at a time and is usually more secure, but

slower to execute than stream encryption.
fe
ea

Task 14C-1 List and describe some of the Electronic Cookbook (ECB) encrypts each block by itself.
Question 3 modes used in block cipher
encryption. Cipher Block Chaining (CBC) encrypts each block with
information from the preceding block.
Pl
Re

Cipher Feedback Mode (CFB) encrypts partial blocks

rather than full blocks.
Output Feedback Mode (OFB) encrypts partial blocks, but
shift registers are used in a different way.

Solutions 495
LESSON 14 ANSWERS
Task/Step Question Response

ial
Task 14C-1 How do DES, RSA, and Digi- DES uses a symmetric key, RSA uses a very large prime
Question 4 tal Signatures encrypt data? number as the public encryption key, and Digital Signa-
tures use a public key to verify the message and a private
key to create the signature.

er
Task 14C-1 Which encryption method Private key uses symmetric and public-key uses
Question 5 uses symmetric key cryptog- asymmetric.
raphy and which uses
asymmetric key cryptogra-
phy?

at
Task 14C-1 What is another name for Diffie-Hellman encryption.
Question 6 public key encryption?

y
Task 14D-1 Packet filtering or screening Data Link, Network, and Transport layers.

op
Question 1 routers.
eM
Task 14D-1 Circuit gateways or circuit Session layer.
Question 2 level proxies.

Task 14D-1
Question 3
Address translation feature of
IP gateways.
Network layer.

tC
No
Task 14D-1 Application gateways or Application level.
Question 4 application-level proxies.
c

Task 14D-2 Specialized network devices Firewalls.

Question 1 that are used to limit access
en

from outside networks to

Do

your company’s network.

Task 14D-2 Servers that isolate internal Proxy servers.

Question 2 network computers from the
Internet.
se
r

Task 14D-2 The most secure firewalls that Application gateways or application-level proxies.
fe

Question 3 operate in the upper layers of

ea

the OSI model.

Task 14D-2 Accelerates Web access per- Proxy cache server.

Question 4 formance, reduces WAN
Pl
Re

traffic and offloads Web

server requests from the Web
server.

496
LESSON 14 ANSWERS
Task/Step Question Response

ial
Task 14D-2 Places multiple copies of the Hierarchical proxy cache servers or network accelerators.
Question 5 cached content on second-
level or peer servers.

Task 14E-1 Identical operations are per- Disk duplexing.

er
Question 1 formed on two disk drives,
each on a separate disk
channel.

Task 14E-1 Enables NetWare to monitor UPS monitoring.

at
Question 2 an uninterruptible power
supply so that NetWare can
alert users and shut down

y
properly if there is a power
outage.

op
Task 14E-1
eM
Blocks of data written to one Disk mirroring.
Question 3 disk are simultaneously writ-
ten to another disk on the

tC
same channel.

Task 14E-1 Backs out transactions that Transaction Tracking System.

Question 4 have been interrupted by the
No
failure of a network
component.
c

Task 14E-1 Blocks from a disk’s main Sector sparing or Hot Fix.
Question 5 storage area are redirected to
en

another portion of the disk.

Do

Task 14E-1 After a block of data is writ- Read-after-write verification.

Question 6 ten to a hard disk, it is read
back from the hard disk and
compared to the original data
se
r

in memory.
fe

Task 14E-1 Prevents a damaged server Duplicate FATs and DETs.

ea

Question 7 table from making data

inaccessible.

Task 14E-2 Which RAID level is also RAID level 1.

Pl
Re

Question 1 referred to as disk mirror-

ing?

Task 14E-2 Which RAID level is also RAID level 0.

Question 2 referred to as disk striping
without parity?

Solutions 497
LESSON 14 ANSWERS
Task/Step Question Response

ial
Task 14E-2 Which RAID level should you RAID level 5. Parity data is spread across drives. Both
Question 3 use for the greatest level of data and parity are spread across drives, so more than one
fault tolerance? Why? read and write can occur simultaneously.

Task 14E-2 Which RAID level would best Answers will vary.

er
Question 4 suit your needs? Why?

Task 14E-2 How many disks are required At least three disk drives and up to 32 disk drives.
Question 5 for RAID level 5?

at
Task 14E-3 What utility is used to create Disk Administrator.
Question 1 a volume set?

y
Task 14E-3 What is the minimum num- 1.
Question 2 ber of disks required to

op
create a volume set?
eM
Task 14E-3 Can a volume set contain a No.
Question 3 boot or system partition?

Task 14E-3
Question 4
Can you resize a volume set?

tC
Yes. You can extend the volume set to enlarge it.
No
Task 14E-4 Name one benefit stripe sets Stripe sets can improve I/O performance.
Question 1 have over volume sets.
c

Task 14E-4 Name two restrictions on For stripe sets, you must have at least two drives, and the
Question 2 stripe sets that don’t apply to free space on each must be about the same size.
en

volume sets.
Do

Task 14E-4 Describe a scenario where Answers might include: You’re using a Windows NT
Question 3 you might implement a stripe computer as a file server for a workgroup. It has multiple
set. hard disks and you want excellent I/O performance.
se
r

Task 14F-1 How many tape sets are One set each for Monday through Thursday (4), One set
Question 1 required when using the for each Friday of the month (5 since some months could
fe

grandfather rotation method? contain 5 Fridays), and one month end set (for the last
ea

day of the month), for a total of 10.

Task 14F-1 How many tape sets are This method requires 5 sets since the rotation schedule is
Question 2 required when using the based on rotating sets A through E.
Pl
Re

Tower of Hanoi rotation

method?

498
LESSON 14 ANSWERS
Task/Step Question Response

ial
Task 14F-1 List and describe the Full—all information is backed up.
Question 3 NetWare backup types.
Incremental—new files and files created or modified since
the last full or incremental backup are backed up and the
archive bit is cleared.

er
Differential—all files created or modified since the last
full backup are backed up and the archive bit isn’t
cleared.

Task 14F-1 List and describe the NT Full/Normal—backups up selected files and marks them

at
Question 4 backup methods. as backed up.
Copy —backs up selected files without marking them as
backed up.

y
Incremental—backs up only selected files that have

op
eM changed since the last backup, and marks them as backed
up.
Daily copy—backs up any selected files that have been

tC
modified that day, without marking them as backed up.
Differential—backs up only selected files that have
changed since the last backup, without marking them as
backed up.
No
c

Topic Review Questions Response

en

What are the levels of secu- Login, file system, and NDS security.
A
Do

rity in a NetWare network?

What are the levels of secu- User-level and share-level security.

rity in a Windows
environment?
se
r

Describe the features of a It should be hard for someone to guess; it should be com-
B
fe

good password. posed of alphanumeric and special character; it shouldn’t

ea

be based on dictionary words or names of people, pets, or

places.

Why might you want to If the data passes through unsecured data communication
Pl

C
Re

encrypt data? lines (such as in a WAN link) or to outside agents (other

companies), it is a good idea to encrypt the data so that if
someone other than the intended recipient gets the data,
they won’t be able to use it.

Solutions 499
LESSON 14 ANSWERS
Topic Review Questions Response

ial
Compare firewalls and proxy Firewalls limit access from outside networks to your
D servers. network. Proxy servers isolate your network computers
from the Internet.

List some of the system fault- Disk mirroring, disk duplexing, sector spanning (Hot Fix),
E

er
tolerance features available UPS monitoring, and for NetWare read-after-write verifi-
on NT and NetWare cation, TTS, and duplicate FATs and DETs.
networks.

Which RAID levels are NT Server includes RAID levels 0,1, and 5. NetWare

at
included with NT Server and includes RAID level 1.
which are included with
NetWare?

y
List some of the backup Magnetic tape (including reel-to-reel, DAT, DLT, and
F

op
eM
media choices available today. QIC)
Optical drives
Recordable CDs

tC
Additional drives and cartridges (such as Iomega Jaz and
Zip disks)
No
c

LESSON 15 ANSWERS
en
Do

Task/Step Question Response

Task 15A-1 Explain the types of network You need an administrative account to be able to install
Question 1 accounts you need for a net- the NOS and to set up the network environment. You
se

work installation and the uses need at least one user or test account to be able to verify
r

for each type. that users will be able to access the correct network
resources.
fe
ea

Task 15A-1 Why should you gather infor- To avoid assigning duplicate addresses, no matter what
Question 2 mation about network type of network protocol you are running. In the case of
addressing before you install TCP/IP networks, to assign the correct subnet mask and
a network component? to determine the addresses of important network
Pl
Re

resources, such as servers and routers.

Task 15A-1 What are some network Answers might include: naming standards, accessibility
Question 3 issues that might be covered standards, restrictions on Internet or email use, and levels
by a company’s SOPs? of administration.

500
LESSON 15 ANSWERS
Task/Step Question Response

ial
Task 15A-1 You are asked to create sev- a. Copies of company SOPs and a list of avail-
Question 4 eral new user accounts on an able network addresses.
existing network prior to per-
b. A valid IP address and subnet mask.
forming a network upgrade.
Which of the following best ✓ c. Copies of company SOPs and an administra-

er
describes the minimum infor- tor account.
mation you need to be able to d. The users’ full names and department num-
perform this task, and why? bers, and the network resources they need to
access.
e. An administrator account and a list of valid

at
IP addresses.

Task 15A-1 You are a consultant working Probably not. The liaison has provided some things you

y
Question 5 with a manufacturing com- don’t need, while omitting some things you do need.

op
pany that wants to upgrade Superfluous items include the list of employees and the
eM
the NOS running on the cor- production documentation. Missing items include network
porate network. The documentation and SOPs related to the network. (Test
company’s liaison gives you a accounts and passwords are also not included, but you’d

tC
project notebook containing a be able to create them with the administrative account.)
temporary security badge, a
comprehensive list of
employee names and phone
numbers, the name and pass-
No
word for an administrative
c

account, a detailed floor map,

documentation for the pro-
duction process, and the
en

naming standards documents

Do

for the current version of the

NOS. Will you be able to per-
form the upgrade
successfully? Why or why
not?
se
r

Task 15B-1 List some environmental con- Responses should include: heat, humidity, and dust.
Question 1 ditions you should check
fe

before installing network

ea

hardware.
Pl
Re

Solutions 501
LESSON 15 ANSWERS
Task/Step Question Response

ial
Task 15B-1 Explain ESD and how you ESD stands for electrostatic discharge and is the transfer-
Question 2 can prevent it from affecting ence of electrons that takes place between objects that
computer equipment. have dissimilar charges. To prevent ESD from damaging
computer equipment, follow ESD precautions, such as
using an ESD kit and static-shielding bags; keeping com-

er
puter components away from metallic and other
conductive sources as well as nonconductive sources such
as plastic, Styrofoam, synthetics, and silks; and maintain-
ing relative humidity levels at 40% to 60%.

at
Task 15B-1 Why should you keep radios Radios and TVs can cause power fluctuations, and are
Question 3 and TVs away from com- generators of radio frequency interference, which can
puter equipment? cause data corruption.

y
Task 15B-1 You are reviewing a site for Advise personnel that any personal electronic devices in

op
Question 4
eM
an upgrade installation. This
company is quite casual,
their cubicles should be placed as far away from their
workstations as possible, and warn them that animal hair
allowing employees to dress and dander can cause their computer equipment to over-
and to decorate their cubicles heat because the particles prevent air from circulating

tC
any way they want, and even within the computers.
allowing them to bring their
pets to work with them.
What suggestions would you
make to help improve the
No
success of the upgrade?
c

Task 15B-1 Which of the following loca- a. In an enclosed wiring closet next to the main
Question 5 tions would be your choice telephone switch panel.
en

for placing a file server, and

Do

✓ b. In a ventilated office with a southwest view.

why?
c. In a corner of the plant floor.
d. In an air-conditioned room that contains
other electronic devices.
se
r

Task 15C-1 Compare and contrast A repeater connects network segments, amplifying the
Question 1 repeaters and hubs. signal so that you can extend the maximum distance of a
fe

cable segment without signal loss. A hub connects work-

ea

stations in a star topology, and, depending on the type of

hub, might provide only a physical connecting point for
the cables (passive), might act as a multiport repeater
(active), or might provide configuration and management
Pl
Re

options (intelligent). Some hubs act as repeaters; others

do not. Both hubs and repeaters can connect different
media types, such as coax and UTP.

502
LESSON 15 ANSWERS
Task/Step Question Response

ial
Task 15C-1 List some network compo- Some print servers, serial peripherals, parallel peripherals,
Question 2 nents that use DB-25 and SCSI devices.
connectors.

Task 15C-1 In what order would the fol- a. Workstation, NIC, patch cable, wall jack, net-

er
Question 3 lowing components usually be work segment, patch panel, hub, network
connected? Explain your backbone, hub, patch cable, patch panel,
response. patch cable, file server.
b. Workstation, UPS, patch cable, wall jack, net-
work segment, patch panel, patch cable,

at
router, network backbone, router, patch cable,
patch panel, patch cable, file server.
✓ c. Workstation, NIC, patch cable, wall jack, net-

y
work segment, patch panel, patch cable, hub,

op
eM network segment, bridge, network segment,
patch panel, patch cable, file server.
✓ d. Workstation, NIC, patch cable, hub, patch
cable, wall jack, network segment, patch

tC
panel, network backbone, hub, patch cable,
file server.

Task 15C-1 Which of the following sets of a. BNC, DB-25, and Centronics-50.
No
Question 4 connectors can be used for
b. Centronics-36 and Centronics-50.
c

SCSI devices? Explain your

response. c. DB-9 and Centronics-50.
✓ d. DB-25 and Centronics-50.
en
Do

Apply Your BNC connector. Pictures of BNC connectors can be found in Figure 4-7 of
Knowledge 15-1 this manual and in the Cables and Connectors topic of the
Question 1 Network Cables and Devices course found in the Digital
Video component of the Network+ Certification
CD-ROM.
se
r

Apply Your Bridge. Pictures of bridges can be found in the Bridges, Routers,
Knowledge 15-1 and Brouters topic of the Network Cables and Devices
fe

Question 2 course found in the Digital Video component of the Net-

ea

work+ Certification CD-ROM.

Apply Your Brouter Pictures of brouters can be found in the Bridges, Routers,
Pl
Re

Knowledge 15-1 and Brouters topic of the Network Cables and Devices
Question 3 course found in the Digital Video component of the Net-
work+ Certification CD-ROM.

Solutions 503
LESSON 15 ANSWERS
Task/Step Question Response

ial
Apply Your Centronics connectors. Pictures of Centronics connectors can be found in the
Knowledge 15-1 TechEncyclopedia at www.techweb.com (search for
Question 4 Centronics).

Apply Your DB-9 connector. Pictures of DB-9 connectors can be found in the

er
Knowledge 15-1 TechEncyclopedia at www.techweb.com (search for serial
Question 5 port).

Apply Your DB-25 connector. Pictures of DB-25 connectors can be found in Figure 6-3
Knowledge 15-1 of this manual.

at
Question 6

Apply Your Hub. Pictures of hubs can be Digital Video component of the

y
Knowledge 15-1 Network+ Certification CD-ROM.
Question 7

op
Apply Your IBM data connector.
eM Pictures of IBM data connectors can be found in Figure
Knowledge 15-1 6-3 of this manual and in the Cables and Connectors
Question 8 topic of the Network Cables and Devices course found in

tC
the Digital Video component of the Network+Certification
CD-ROM.

Apply Your Network interface card. Pictures of network interface cards can be found in Figure
No
Knowledge 15-1 6-2, 6-9, 6–11 and 6-12 of this manual.
Question 9
c

Apply Your Patch panel and patch cable. A picture of a patch panel and cables can be found in
Knowledge 15-1 Figure 4-4 of this manual.
en

Question 10
Do

Apply Your Print server. Pictures of print servers can be found on manufacturers’
Knowledge 15-1 Web sites such as www.hp.com.
Question 11
se
r

Apply Your Repeater. Pictures of repeaters can be found in the Network

Knowledge 15-1 Devices topic of the Network Cables and Devices course
fe

Question 12 found in the Digital Video component of the Network+

ea

Certification CD-ROM.

Apply Your RJ-11 connector. Pictures of RJ-11 connectors can be found in the Cables
Knowledge 15-1 and Connections topic of the Network Cables and
Pl
Re

Question 13 Devices course found in the Digital Video component of

the Network+ Certification CD-ROM.

504
LESSON 15 ANSWERS
Task/Step Question Response

ial
Apply Your RJ-45 connector. Pictures of RJ-45 connectors can be found in Figure 4-3
Knowledge 15-1 and 5-8 of this manual and in the Cables and Connections
Question 14 topic of the Network Cables and Devices course found in
the Digital Video Component of the Network+ Certifica-
tion CD-ROM.

er
Apply Your Router. Pictures of routers can be found in the Bridges, Routers,
Knowledge 15-1 and Brouters topic of the Network Cables and Devices
Question 15 course found in the Digital Video component of the Net-
work+ Certification CD-ROM.

at
Apply Your SC connector. Pictures of SC connectors can be found in the Cables and
Knowledge 15-1 Connections topic of the Network Cables and Devices

y
Question 16 course found in the Digital Video component of the Net-
work+ Certification CD-ROM.

op
Apply Your SCSI connectors.
eM Pictures of SCSI connectors can be found in the
Knowledge 15-1 TechEncyclopedia at www.techweb.com (search for
Question 17 SCSI).

Apply Your
Knowledge 15-1
Question 18
ST connector.

tC
Pictures of ST connectors can be found in the Cables and
Connections topic of the Network Cables and Devices
course found in the Digital Video component of the Net-
No
work+ Certification CD-ROM.
c

Apply Your UPS. Pictures of UPSs can be found on manufacturers’ Web

Knowledge 15-1 sites such as www.apc.com.
Question 19
en
Do

Apply Your USB connectors. Pictures of USB connectors can be found in the
Knowledge 15-1 TechEncyclopedia at www.techweb.com (search for USB).
Question 20

Task 15D-1 Before you connect a new a. Make sure that the wall jack has two connec-
se
r

Question 1 modem to the telephone sys- tions so that a phone can also be connected.
tem, which of the following
b. Make sure that the modem is placed within
fe

should you do? Explain your

arm’s length of the wall jack.
ea

response.
✓ c. Make sure that the modem is compatible with
the workstation.
✓ d. Make sure that the wall jack is connected to
Pl
Re

an analog phone line.

Solutions 505
LESSON 15 ANSWERS
Task/Step Question Response

ial
Task 15D-1 You’re running UTP cable in ✓ a. Main cable run.
Question 2 an Ethernet environment
✓ b. Hub patch cable.
from a workstation to a hub,
and then to a wall jack. ✓ c. Workstation patch cable.
Which of the following are d. Hub internal wiring.

er
included in the overall cable
length? Explain your
response.

Task 15D-1 You are a consultant who has First, ask the user if he has placed any service requests

at
Question 3 been asked to replace a work- regarding the old computer (that might be why he’s get-
station in the Customer ting the new machine). Then, record all the issues he
Service Department. When mentions, including any error messages he receives when

y
you arrive in the department trying to access local or network resources. List the appli-
with the new computer, the cations installed on the old machine, and verify that they

op
plaints about the old
eM
user begins a litany of com- are to be installed on the new machine. Make arrange-
ments for the user to back up data files so that they can
computer. What should you be transferred to the new workstation. Also, make a note
do before replacing the work- to test the new machine and verify user satisfaction

tC
station? before closing the service request.

Topic Review Questions Response

No
List some information you Administrative and test accounts with passwords, network
A
c

should gather before install- addressing information, and any company SOPs that
ing a new or upgraded apply to network usage.
network operating system.
en
Do

Describe a good environment The ideal computer environment would be an enclosed,

B for computers. climate-controlled area, with a temperature level of about
70° F. and humidity between 40% and 60%. All comput-
ers would sit on ESD mats, would be connected to UPSs,
and would be situated as near as possible to their electri-
se

cal and data outlets. No portable electronic devices would

r

be allowed in the area.

fe

What is an advantage of Many UPSs contain software to initiate a safe shutdown

ea

C using UPSs with file servers? in case of long power outages.

What types of compatibility Cable, connection, hardware, and software compatibility.

Pl
Re

D issues should you review

before performing an installa-
tion or upgrade?

506
LESSON 16 ANSWERS
Task/Step Question Response

ial
Task 16A-1 Where’s the first place you In the product packaging.
Question 1 should look for vendor docu-
mentation?

Task 16A-1 What types of information FAQs, white papers, utilities, support and troubleshooting

er
Question 2 might you find on a subscrip- issues, and software updates and patches.
tion CD?

Task 16A-1 Welcome to Microsoft Prod- http://www.microsoft.com/support

Question 3 uct Support Services

at
Microsoft TechNet Support http://www.microsoft.com/technet/support/default.htm

Microsoft TechNet CD Online http://technet.microsoft.com/cdonline/default.asp

y
Site - Home

op
eM
Novell Customer Services http://services.novell.com

Novell: Product Documenta- http://www.novell.com/documentation

tC
tion

Novell Support Connection http://support.novell.com

The Novell Support Connec- http://support.novell.com/additional/nsc-cd.htm

No
tion CD
c

Novell: NetWare Users Inter- http://www.novell.com/corp/community/nui/index.html

national
en
Do

Cisco Technical Assistance http://www.cisco.com/pcgi-bin/ibld/all.pl?i=support&c=2&m=GUES

Center

CNET Help.com http://www.help.com

se
r

SupportSource http://www.supportsource.com
fe

ZDNet: Help & How-To http://www.zdnet.com/zdhelp/filters/home/0,8042,6006925,00.html

ea

Task 16B-1 What are some benefits of Responses should include the ability to recover lost data
Question 1 having a good backup plan? without substantial waiting periods and confidence that
network data is safe from loss.
Pl
Re

Solutions 507
LESSON 16 ANSWERS
Task/Step Question Response

ial
Task 16B-1 Which backup type requires a. Full.
Question 2 no more than two backup
b. Partial.
sessions to fully recover data?
Explain your answer. ✓ c. Differential.
d. Incremental.

er
Task 16B-1 Which of the following cor- a. Magnetic tape, removable optical disks, large-
Question 3 rectly lists backup media in capacity nonremovable disks, small-capacity
order of diminishing capac- removable disks.
ity? Explain your response.
b. Magnetic tape, large-capacity removable

at
disks, removable optical disks, small-capacity
removable disks.

y
c. Small-capacity removable disks, large-
capacity removable disks, removable optical

op
eM disks, magnetic tape.
✓ d. Magnetic tape, removable optical disks, large-
capacity removable disks, small-capacity

tC
removable disks.

Task 16B-1 At least monthly, backup a. Scanned for viruses.

Question 4 tapes should be:
b. Replaced.
No
✓ c. Retained.
c

d. Taken offsite.
e. Created.
en
Do

Task 16B-1 What recording method is Helical scan.

Question 5 used to record on DAT tape?

Task 16B-1 You get a call on Tuesday First, verify that the spreadsheet was not saved on her
Question 6 from a user who can’t access workstation by asking her to search her hard drive for the
se
r

the budget spreadsheet she file. Next, determine the last day that she worked with the
saved to the network. She file, and compare that date with your backup schedule to
apologizes for not calling determine which backup set is most likely to contain the
fe

sooner, but she just returned file. Then, obtain the backup set and restore the file to the
ea

from a long weekend. How network. Finally, verify with the user that she can open
can you help? and work with the restored file.
Pl

Task 16C-1 What are the two components Engine and definition files.
Re

Question 1 of anti-virus software that

you can update to ensure a
proper level of virus protec-
tion?

508
LESSON 16 ANSWERS
Task/Step Question Response

ial
Task 16C-1 If you can’t save a Microsoft a. Your computer is infected with a boot-sector
Question 2 Word document, what’s the virus. You’ll have to format the hard drive
most likely cause and solu- and reinstall all software.
tion? Explain your response.
b. Your computer is infected with a worm virus.
You’ll have to scan and clean all viruses from

er
the system.
c. Your computer is infected with a boot-sector
virus. You’ll have to copy the file into
WordPad and restart Word.

at
✓ d. Your computer is infected with a macro virus.
You’ll have to scan and clean all viruses from
the system.

y
Task 16C-1 List at least two ways of Manually downloading them from the Internet, and

op
Question 3
eM
obtaining updated virus sig- Internet push technology.
nature files.

Task 16C-1 A user downloads a freeware a. The proposed solution meets the required

tC
Question 4 program from the Internet results.
onto a Windows NT
b. The proposed solution meets the required
Workstation. The next day,
results and one of the optional results.
when he tries to start the
No
computer, the Missing Oper- ✓ c. The proposed solution meets the required
c

ating System message is results and both of the optional results.

displayed. This computer has d. The proposed solution doesn’t meet the
the latest virus scanning soft- required results.
en

ware and definition files

Do

installed. You must get the

computer working properly
and get rid of the virus.
Optionally, you also want to
ensure that the computer is
se

protected from downloaded

r

files and viruses. A colleague

suggests that you reboot the
fe

computer in Safe Mode, then

ea

run an on-demand scan and

clean all files, reconfigure the
virus scanning software to
scan and clean all down-
Pl
Re

loaded files, and reboot the

computer.

Task 16D-1 Where’s the best place to get From the vendor or a reseller.
Question 1 patches and updates for your
NOS?

Solutions 509
LESSON 16 ANSWERS
Task/Step Question Response

ial
Task 16D-1 Other than NOS patches and Device drivers.
Question 2 application updates, what
other types of files need to be
monitored for updates?

er
Task 16D-1 You have to install a patch a. A test machine, your workstation, manage-
Question 3 for the spreadsheet applica- ment workstations, production workstations.
tion used by the entire
✓ b. A test machine, your workstation, support
company. In what order
staff workstations, selected users’ worksta-
should this patch be applied?
tions, production workstations.

at
Explain your response.
c. Your workstation, selected users’ worksta-
tions, production workstations.

y
d. A test machine, your workstation, selected
users’ workstations, production workstations.

op
eM
Task 16D-1 How often should you check At least quarterly.
Question 4 for updates to network oper-
ating systems and

tC
applications?

Task 16D-1 How long should you test an At least two weeks.
Question 5 OS patch before rolling it out
No
to production servers?
c

Topic Review Questions Response

en

What is the best source of The vendor’s Web site.

Do

A free information about a net-

working product?

List some information that Responses should include: the types of data that need to
B should be included in a be backed up, the backup media to be used, the timing
se
r

backup plan. and frequency of backup operations, the person or group

responsible for performing and maintaining the backup
fe

schedule, where the backup media should be stored, a

ea

schedule for testing the backups to ensure their validity,

and a plan for recovering data from backups.

Where’s the best place to On all servers and workstations.

Pl

C
Re

install anti-virus software?

How often should you update At least monthly, and immediately after you receive a
virus definition files? new virus alert.

510
LESSON 16 ANSWERS
Topic Review Questions Response

ial
How can you determine By reviewing the documentation describing each patch,
D which NOS patches you determining if the patch will address any current prob-
should apply to your net- lems you’re experiencing, and testing the patch to ensure
work? that it doesn’t cause other problems on your network.

er
LESSON 17 ANSWERS

at
Task/Step Question Response

y
Task 17A-1 How can you determine the See if you can replicate the problem on another computer.
Question 1 scope of a reported network

op
problem?
eM
Task 17A-1 After you isolate the cause of Formulate a correction plan.
Question 2 a network problem, what’s

tC
the next step?

Task 17A-1 Why is it important to docu- To save time if the same or a similar problem arises in
Question 3 ment network problems? the future.
No
c

Task 17A-1 What are some quick fixes Responses might include: verifying that the client soft-
Question 4 you can try when a user com- ware is loaded on the workstation, that the user is using
plains that she can’t connect the correct login name and password, and that the work-
en

to the network? station is physically connected to the network.

Do

Task 17A-1 What’s the most often over- Verifying user satisfaction.
Question 5 looked part of a
troubleshooting plan?
se

Task 17A-1 What do all of the models They all provide a systematic approach to solving prob-
r

Question 6 discussed have in common? lems, and they all include steps for collecting information
about the problem, isolating the cause of the problem, and
fe

implementing a solution to the problem.

ea

Task 17A-1 Users on the third floor a. Router configuration tables.

Question 7 report that they can’t connect
b. If viruses exist on the network.
Pl

to the Internet, but they can

Re

connect to the NetWare c. Power cable to the hub.

server. What would you ✓ d. If users on other floors are having similar
check first, and why? problems.

Solutions 511
LESSON 17 ANSWERS
Task/Step Question Response

ial
Task 17B-1 Users throughout the com- a. Protocol analyzer and crossover cables.
Question 1 pany are complaining that
✓ b. Server console screen and protocol analyzer.
network response times are
too slow. Which of the follow- c. Performance monitors and cable testers.
ing should you use to help d. Log files and performance monitors.

er
determine the cause, and
why?

Task 17B-1 What’s the primary use for a To perform diagnostic tests on NICs.
Question 2 loopback?

at
Task 17B-1 What tool would you use to Responses might include winipcfg and protocol analyzers.
Question 3 solve IP addressing prob-

y
lems?

op
Task 17B-1
Question 4
eM
When is the most likely time
to use a tone generator and
When you need to trace one cable in a bundle or group of
cables.
tone locator?

tC
Topic Review Questions Response

List the steps of the Net- Identify the exact issue. Re-create the problem. Isolate the
A work+ troubleshooting model. cause. Formulate a correction. Implement the correction.
No
Test the solution. Document the problem and the solution.
c

Provide feedback.

What types of network prob- Physical layer problems.

B
en

lems can hardware

Do

troubleshooting tools help you

detect?
se
r
fe
ea
Pl
Re

512
 GLOSSARY

ial
(CRC) Cyclic Redundancy Check Application Programming Interface (API)
An error-detection method used to detect Application Programming Interface. An
data corruption due to transmission errors. extension to a programming language that
defines an interface to an operating system.
56 K line If an application is written for a particular
Switched or dedicated 56 K connection API, it need only be recompiled to run on
using a partial T1 line.

er
any other operating system that provides
that API.
abend
Short for ″abnormal end.″ A condition that ARIN
prevents a NetWare server from working American Registry for Internet Numbers.
properly. Regional Internet registry for obtaining IP

at
addresses in the United States.
active hub
Hubs that regenerate and retransmit the ARP

y
signal. Address Resolution Protocol. The Internet

op
layer protocol that translates software
address-to-name resolution
eM
addresses to hardware addresses.
The process of mapping an alphanumeric
address to an understandable name. asynchronous

tC
A bit synchronization transmission tech-
admin
nique that uses start and stop bits.
The superuser on NetWare 4.x and higher
networks. This user has full administrative ATM
rights to the Directory tree objects and to Asynchronous Transmission Method. High-
No
the file system. speed packet-switched technology using
c

53-byte cells and error-checking.

Administrator
The superuser account on an NT system. attenuation
en

Degradation of signal.
anti-virus engine
Do

An executable file that uses virus definition AUI connector

or virus signature files to identify the pres- (Attachment Unit Interface) A 15-pin
ence of viruses on a computer and D-shaped connector (a DB-15 connector)
eradicates the viruses it finds. used in thicknet cable networks.
se
r

APNIC AWG
Asia Pacific Network Information Center. A measurement of wire diameter, where the
fe

Regional Internet registry for obtaining IP diameter of the wire increases as the AWG
ea

addresses in the Asia-Pacific region. number gets smaller. Thus, AWG 22 wire
is thicker than AWG 26 wire.
Application layer
Level 7, the highest layer in the OSI backbone
Pl
Re

model. It provides a user interface. The main trunk of a network communica-

tion channel.

Glossary 513
 GLOSSARY

ial
bandwidth chip creep
The difference between the highest and A condition caused by repeated moving of
lowest frequencies used for a communica- computer equipment to different
tion channel. Generally, more bandwidth temperatures. Changes in temperature cause
means greater transmission capacity. the components to expand and contract,
which can in turn cause the connections
baseband

er
between a chip and its socket to become
A signal transmission that uses all of a loose.
channel’s bandwidth.
cipher
BNC A secret message or code.
(British Naval Connector or Bayonet Nut

at
Connector) Connector usually used for Classless Inter-Domain Routing (CIDR)
thinnet coaxial cable. Provides three-way See supernetting.
connection—two connections for the net-

y
work and one for the transceiver device. client-server network
A network in which duties are divided

op
boot PROM
eM between clients, which are users of net-
Chip, used on network adapters in diskless work services, and servers, which are
workstations, that stores the program to providers of network services.

tC
boot the computer, connect the network,
and download the operating system. coaxial cable
A data-transmission medium that contains a
bounded media single-wire inner conductor surrounded by
A physical product, such as cable, through a layer of insulation, a shield of braided
No
which data is transmitted. wire mesh or foil, and an outer layer of
c

insulation.
bridge
A device that connects two LAN segments collision
en

and forwards data to the segment that it’s The loss of data due to signals mixing
Do

addressed to. together and becoming damaged beyond

effectiveness.
broadband
A transmission channel that can carry mul- concentrator
tiple signals simultaneously. A general term for hubs, MSAUs, and
se

switches, all of which concentrate access of

r

brouter multiple devices to a single device.

A communication device, often referred to
fe

as an internetworking device, that functions connection identifier

ea

as both a bridge and a router. Also known as connection ID socket or

port. An address that identifies each line of
bus topology conversation to the upper layers.
A network layout in which nodes are con-
Pl
Re

nected to a single cable or backbone. connection-oriented protocol

A data communication method in which
cellular topology packets are sent and acknowledged by the
A network layout in which nodes commu- recipient and resent if there is any error.
nicate using wireless connections.

514
 GLOSSARY

ial
connectionless protocol data rate
A data communication method in which How fast the data moves within the
frames are ″blindly″ sent over media with- medium.
out knowing if the receiver can receive
them. de facto standard
A standard that exists through popular
Container object practice.

er
An object that can contain other objects
and is used to organize the Directory tree. de jure standard
A standard that exists through codes, laws,
contention decrees, or other forms of legislation.
A media-access method in which stations (Compare with de facto standard.)

at
listen until no other devices are transmit-
ting before they attempt to transmit a default gateway
signal. An IP router that provides access to a

y
remote network.
convergence

op
The synchronizing of all routing tables in a
eM
default gateway address
RIP environment. The network segment’s access point to all
other remote networks. The default gate-
count-to-infinity way address is the IP address of the router

tC
A continuous cycle of hop count recalcula- that the system will use to connect to the
tion in which a packet is never forwarded other remote networks.
to its destination.
demand priority
No
crossover cable A new media-access method for 100VG-
c

A special network cable used in Ethernet AnyLAN networks. A hub determines

UTP installations that enables you to con- when each computer can transmit.
nect two hubs, or to connect two stations
DHCP
en

without using a hub.

Do

Dynamic Host Configuration Protocol.

crosstalk Used to automatically assign or lease IP
Signal overflow from an adjacent wire. addresses to hosts.

CSU/DSU DIX connector

(Channel Service Unit/Data Service Unit) Another name for an AUI connector,
se
r

Combines the functions of CSU and DSU named for the three companies that devel-
in one device. The CSU terminates a digi- oped it: Digital, Intel, and Xerox.
fe

tal circuit. The DSU terminates a data

ea

circuit to the Data Terminal Equipment and DNS

converts customer transmission data into a Domain Name System. Resolves host
bipolar format. names to IP addresses.
Pl
Re

Data Link layer domain

Layer 2 in the OSI model. It shares the An organizational unit on a Windows NT
media and controls the flow. network in which workstations and servers
share a single accounts database and are
administered as a group.

Glossary 515
 GLOSSARY

ial
driver gateway
A software program that enables a com- Software or hardware that connects dispar-
puter to communicate with a device. ate types of networks, translating as
necessary between the two.
EMI
Electromagnetic interference. Noise or global group
crosstalk that can severely hinder data Groups of users across multiple NT

er
communication. domains.

encryption half-duplex
The process of converting the data into Data transmission on a channel that permits
coded form in such a way that only autho- communication one way at a time.

at
rized parties can access the information.
hardware loopback plug
external router A special connector used for diagnosing

y
A router that isn’t part of a file server. transmission problems. It plugs into a port
and crosses over the transmit lines to the

op
FAQ
eM receive lines so that outgoing signals can
Frequently Asked Questions. be redirected back into the computer for
testing.
fiber optic cable

tC
A cable that transmits signals of light Hello packet
through a very fine strand of plastic or A means of gathering router information
glass (optical fiber) called the core, which when using OSPF.
is encased in a tube made of glass, called
No
cladding, which in turn, is encased in a hierarchical network
c

tough outer sheath called a jacket. A network in which a host controls net-
work communications and processing.
firewall
host table
en

Specialized network devices used to limit

Do

access from outside networks to your com- An ASCII text file that contains the host
pany’s networks. names and the associated IP addresses for
any host that is commonly accessed on a
Frame Relay TCP/IP network.
A fast-packet technology that uses fiber
optic lines. hub
se
r

A central connecting device in a network

frames that connects communications lines in a
fe

The smallest unit of data sent over the star configuration.

ea

network.
hybrid topology
ftp A combination of two or more different
File Transfer Protocol. A protocol in the topologies.
Pl
Re

Process/Application layer of the DOD

model that enables the transfer of files IBM data connector
between a workstation and a remote host. A device used to connect IBM Token Ring
stations by using Type 1 STP cable.
full-duplex
Simultaneous two-way channel
transmission.

516
 GLOSSARY

ial
ifconfig LAN (Local Area Network)
A UNIX utility that displays the IP con- A collection of computers cabled together
figuration information. to form a network in a small geographic
area (usually within a building).
interconnectivity
The ease and ability of connecting net- LAT
works together. Local Area Transport.

er
internal router latency
A file server that functions as a router. The amount of time it takes a packet to
travel from sender to receiver.
Internet Protocol (IP)

at
The component of the TCP/IP protocol Leaf object
suite that addresses and routes data packets An object that can’t contain other objects,
across the network or internetwork. such as a User object.

y
Internet Protocol suite Link State Advertisement (LSA)

op
A protocol suite that includes TCP and sev-
eM
A signal sent to other routers to update
eral other related data-communication their databases.
protocols.
LLC

tC
internetwork Logical Link Control. Performs Data Link
A network of networks. layer functions related to the Network
layer, such as providing and maintaining
InterNIC the link to the network.
No
Internet Network Information Center. An
local group
c

integrated network information center

developed by Network Solutions, AT&T, Groups of users in a single NT domain.
General Atomics, and the U.S.
logical topology
en

Government. It provides and coordinates

The network path through which data
Do

services for the Internet.

travels.
interoperability
How internetworks communicate and share loopback
resources. Data is sent out and ″loops″ around back
into the card. The ″received″ data is com-
se
r

IPCONFIG pared to the sent data to verify that it’s the

A Windows NT utility that displays the IP same. There are internal loopback tests and
fe

configuration information. those that require an external loopback

ea

plug to test the card.

isochronous
Uses a single network device for clocking MAC
and all other devices set their internal Media Access Control. The portion of the
Pl
Re

clocks to this one device. Data Link layer that controls access to the
communication channel.
jabber
A term describing a condition where a NIC MAN (Metropolitan Area Network)
sends unnecessary packets onto the net- A network linking several buildings within
work, causing excessive traffic. a city or metropolitan area.

Glossary 517
 GLOSSARY

ial
mesh topology NDS
Network architecture in which each node Novell Directory Services. Formerly known
has a dedicated connection to all other as NetWare Directory Services. An admin-
nodes. istrative database that defines network
entities, such as users, groups, servers, and
message so on. NDS was introduced in NetWare
A chunk of data that is transmitted over a

er
4.0, and controls access to network infor-
network. mation, resources, and services. It allows
users to access all network resources with a
metric
single login procedure.
The cost associated with a given route,
usually measured in hops, or the number of NDS (Novell Directory Services)

at
routers that must be used to complete the Formerly known as NetWare Directory
route. Services. An administrative database on the

y
network that defines network entities, such
MIB
as users, groups, servers, and so on. NDS
Management Information Base. A database

op
eM
of information about defined objects, such
was introduced with NetWare 4.0, and con-
trols access to network information,
as physical and network addresses, statis-
resources, and services. It allows users to
tics, counters, and route and ARP tables.
access all network resources with a single

tC
modem login procedure.
MOdulator/DEModulator that changes elec- NetBEUI
tric (digital) signals from a computer to
NetBIOS Extended User Interface.
analog signals for a microwave transceiver
No
or phone line. NetBIOS
c

Network Basic Input/Output System.

multipoint connection scheme
Connection of three or more devices by a NetBIOS Name Server (NBNS)
en

communication channel. Any computer on a TCP/IP network that

Do

performs NetBIOS name resolution. WINS

multiport repeater
is an example of a NBNS.
A hub in a 10BASE-T network is often
known as a multiport repeater, because it NETSTAT
sends any input signal to all outputs. Also A Windows and UNIX utility used to dis-
se

referred to as an active hub.

r

play protocol statistics and current TCP/IP

network connections.
narrow-band radio
fe

A radio signal where both the transmitter network

ea

and receiver are tuned to a certain fre- A collection of hardware and software that
quency to send and receive data. enables a group of computers to communi-
cate and provide users with access to
NBTSTAT
Pl
Re

shared resources.
A Windows utility used to view and man-
age NetBIOS name cache information.

518
 GLOSSARY

ial
network interface card (NIC) patch
A printed circuit board that plugs into both A free update provided by a software
the clients (personal computers or worksta- manufacturer that address known problems
tions) and servers, connecting them to the or provides enhanced services. Also
transmission medium, and controls the referred to as fixes, support packs, and ser-
exchange of data between them. Also vice packs.

er
referred to as a network interface board or
network adapter. patch cable
A short length of cable (usually no longer
Network layer than 10 feet) used to connect ports in patch
Layer 3 in the OSI model. This layer deals panels or in expansion boards and systems
with routing and addressing. that are in close proximity. Also referred to

at
as a patch cord.
network operating system
A collection of software that enables net- patch panel

y
work nodes to communicate with other A group of sockets that acts as a manual

op
nodes. switching center between incoming and
eM
outgoing lines in a communications, elec-
network topology tronic, or electrical system.
The layout of the transmission medium and
peer-to-peer network

tC
devices on a network.
Each workstation on the network acts as
node both a server and a client by sharing its
A network access point. Examples include resources with other computers and
terminals and computers. acessing others’ shared resources.
No
c

one-way trust relationship peripheral

A trust relationship between two domains Any hardware device connected to a com-
where only one domain trusts the other puter, such as a monitor, keyboard, printer,
en

domain. In a one-way trust relationship, the disk, tape, graphics tablet, scanner, joy
Do

trusted domain can access resources in both stick, paddle, and mouse.
domains.
Physical layer
OSPF Layer 1 in the OSI model. It is the physi-
Open Shortest Path First. A link-state rout- cal and electrical interface.
se

ing protocol used on IP networks.

r

physical topology
partition A network’s entire physical configuration.
fe

A logical area of disk space that you can

ea

format and treat as a single storage unit. PING

A utility for checking IP connectivity.
passive hub
Hubs that act only as a connection point plenum cable
Pl
Re

and do not regenerate the signal. Cable made of special materials in the
insulation layers that make it fire resistant.
When it burns, it produces a minimal
amount of smoke and chemical fumes.

Glossary 519
 GLOSSARY

ial
point-to-point connection scheme promiscuous mode
Two devices are connected by a single A condition where a NIC passes all net-
communication channel. work events up the protocol stack to the
operating system and running applications.
poison reverse
An algorithm ensuring that a router broad- protected mode
casts a route cost of 16 for all Program instructions can access only spe-

er
transmissions on its network. cific contiguous areas of the first 640 KB
of RAM. The local operating system typi-
polling cally runs in protected mode to prevent
A media-access method in which a primary data from being overwritten by other
device controls which secondary device has programs. Compare with real mode.

at
access to the data-transmissions media.
protocol
POSIX A code or set of rules by which communi-

y
Portable Operating System Interface for cation is initiated, maintained, and
UNIX. A set of standards developed by

op
terminated.
eM
IEEE, specifying components of a portable,
vendor-independent operating system. The protocol analyzer
goal of POSIX is to enable the develop- Diagnostic software that can examine and

tC
ment of portable application software display data packets that are being trans-
across different vendors’ hardware. mitted over a network.

PPP protocol stack

Point-to-Point Protocol. The Internet stan- A layered set of protocols working together
No
dard for sending IP datagram packets over to provide a set of network functions. Each
c

serial links (such as modem/phone protocol layer services the layer above by
connections). using the layer below.
en

PPTP protocol suite

Do

Point-to-Point Tunneling Protocol. A proto- A set of related protocols from a single

col that connects clients and servers by source.
using Remote Access Services (RAS).
PPTP can be used to create a virtual pri- proxy server
vate connection between computers over Server used to isolate internal network
se

computers from the Internet.

r

the Internet. It is an extension of PPP.

Presentation layer RAID

fe

Layer 6 in the OSI model. It provides code Redundant Array of Inexpensive Disks. An
ea

and syntax. arrangement of multiple disk drives and a

sophisticated controller to offer higher per-
print servers formance or reliability, or both, than a
Pl

A computer or hardware device that con- single disk drive.

Re

trols network print services.

RARP
PROM Reverse Address Resolution Protocol. The
Programmable Read Only Memory. A Internet layer protocol that translates hard-
persistant storage location for data or a ware addresses to software addresses.
program.

520
 GLOSSARY

ial
real mode RJ-45
Program instructions can access any A connector used with twisted-pair cable
address space in the first MB of RAM. that contains eight cable connections.
Usually these programs run in real mode
because they need to access and use system root
information. Compare with protected mode. The superuser on UNIX systems. This is a
special login name used by the system

er
receiver administrator to gain full administrative
The component on the ″hearing″ end of a access to the system. Logging in as root
transmission. enables you to perform any valid command
on any file, regardless of what the permis-
relay agent sions on that file are.

at
Enables hosts to obtain IP addresses from
DHCP servers located on other subnets. [Root] object
Represents the top of the NDS tree and is

y
release notes created during the installation of the
A document that describes the history of a

op
NetWare network operating system.
product, and identifies the changes imple-
eM
mented in specific product versions. router
A device that connects networks and can
repeater

tC
determine the best path for data when mul-
A device that connects two network seg- tiple paths are available.
ments to make them work as one.
Repeaters can extend the length of a net- segment
work beyond the physical limitations of a Secondary communication channel con-
No
single cable. nected to the backbone.
c

resolver service address

A DNS client that submits requests to the Address (number) of a port or socket.
en

local domain server for IP addresses.

Do

service data unit

RFC The data transferred between layers of the
Request For Comments. A numbered series OSI model.
of documents containing information and
standards relating to the Internet. Docu- Session layer
Layer 5 in the OSI model. It provides a
se

ments are posted, then others can review

r

and comment on them. logical connection.

fe

ring topology shielded twisted pair

ea

A network topology that connections net- Twisted pair wire that’s surrounded by a
work devices in a continuous loop. foil or braided metal sheath for the purpose
of reducing susceptibility to electromag-
RIPE netic interference.
Pl
Re

Reseau IP Europeens. Regional Internet

registry for obtaining IP addresses in simplex transmission
Europe. One-way transmission.

Glossary 521
 GLOSSARY

ial
SLIP star topology
Serial Line Internet Protocol. An older A network topology in which nodes are
method for sending IP bytestreams over connected to a central hub.
serial lines (such as modem/phone
connections). stripe set
The combination of 2 to 32 areas of free
SMDS space on 2 to 32 physical disks into one

er
Switched Multimegabit Data Service. Simi- logical drive.
lar technology to ATM, but uses no error
correction. subnet
An independent network physically con-
SNA nected to and part of an internetwork.

at
Simple Network Architecture.
Subnet mask
SNMP A 32–bit address used to mask the host

y
Simple Network Management Protocol. portion of the IP address. It is also used to
The Internet standard protocol developed to subdivide a single IP address into multiple

op
eM
manage nodes on an IP network. It can be subnets.
used to monitor and network devices and
their functions. Supernetting
Bits are borrowed form the network ID and

tC
SONET masked as the host ID. Used to help pre-
Synchronous Optical Network. Uses fiber vent depletion of network addresses.
optic media for high-speed
communications. superuser
No
A network account with full access and
c

SOP rights to everything on the network in

Standard Operating Procedure. SOPs out- order to administer it.
line the basic administrative workings of a
SUPERVISOR
en

company, from hiring guidelines to rules

Do

controlling the use of company resources The superuser account on NetWare 3.x and
during non-work hours. lower networks. This user has full adminis-
trative rights to create users, groups, print
spanning tree protocol services, and to administer the file system.
Arrangement of segment routes and redun-
synchronous
se

dant bridge routes to prevent brigding

r

loops. Transmission of a bit stream of data where

the transmitter and receiver are
fe

split horizon synchronized.

ea

An algorithm that prevents a router from

broadcasting internal network information. TCP/IP
A group of protocols that specify how dis-
spread-spectrum radio similar computers can be connected.
Pl
Re

A radio signal where the transmitter and

receiver tune into a specific channel for a
predetermined period of time. They then
switch to a different channel periodically
until the data has been transmitted and
received.

522
 GLOSSARY

ial
telnet Transmission Control Protocol (TCP)
A terminal emulation protocol in the The component of the TCP/IP protocol
Process/Application layer of the DOD suite responsible for establishing a connec-
model that enables a user connect to a host tion between the sender and the receiver in
computer over the network and run a ter- a network transmission. It guarantees deliv-
minal session. ery of the data through acknowledgments

er
and timers.
terminator
A resistor attached to the end of the cable transmission media adapter
to prevent signal reflections, making the Receives signals from one of the connec-
cable appear to be infinitely long to the tors and converts them to a form
signals being sent across it. interpretable by the other connector.

at
throughput transmission media connector
A measure of productivity on a device or Joins a network device to a media cable.

y
network.
transmitter

op
token passing
eM
The component on the ″speaking″ end of a
A media-access method in which the sta- transmission.
tion with the token is the only one who can
Transport layer

tC
transmit.
Layer 4 in the OSI model. This layer is the
tone generator heart of the model; it provides reliability.
An electronic device that sends an electri-
cal signal through one set of UTP wires. Trojan horse
No
Sometimes referred to as a fox. A program that has a different purpose than
c

expected or stated.
tone locator
An electronic device that emits an audible trusted domain
en

tone when it detects a signal in a set of A domain that is part of a trust relationship
Do

wires. Sometimes referred to as a hound. with another domain. When a domain is

trusted, the trusted domain’s users can be
TRACERT granted permissions to other domains’
A utility used to see the route data uses to resources.
get to its destination.
trusting domain
se
r

transaction identifier A domain that is part of a trust relationship

Enables the service provider to handle the with another domain. When a domain is
fe

necessary tasks for a request. Similar to trusting, it makes its resources available to
ea

connection ID, but works on transactions users from other domains.

rather than a whole conversation. Also
known as transaction ID. twisted pair
Cable consisting of at least two insulated
Pl
Re

transceiver wires that are intertwined to reduce electro-

A device that can function as a transmitter magnetic interference.
or receiver.

Glossary 523
 GLOSSARY

ial
two-way trust relationship white paper
A trust relationship between two domains A detailed technical report written by a
where both domains trust each other. In a vendor, consultant, research entity, or other
two-way trust relationship, users from both organization that describes a specific topic
domains can access resources in either of interest.
domain.
WINIPCFG

er
uninterruptible power supply (UPS) A Windows 95/98 utility that displays the
A device that provides backup power when IP configuration information.
the electrical power fails or drops to an
unacceptable voltage level. WINS
Windows Internet Name Service. Automati-

at
unshielded twisted pair cally maps NetBIOS (computer) names for
Twisted pair wire that isn’t protected from IP addresses in commands and utilities.
EMI through the use of shielding.

y
X.25
virus Switched virtual circuits used to dynami-

op
eM
A program that runs without your knowl- cally connect remote terminals to
edge or permission, often causing damage mainframes.
to your computer or data, and that replicate

tC
themselves so that they can be spread to
other computers.
virus definition file
A file that lists known viruses and how to
No
clean them from computers. Generally
c

updated on a monthly basis. Also referred

to as virus signature file.
en

virus signature
Do

The binary pattern of the machine code of

a particular virus. Anti-virus engines com-
pare their database of virus signatures with
the files on the hard disk and removable
media (including the boot sectors of the
se
r

disks) as well as within RAM.

volume set
fe

The combination of 2 to 32 areas of free

ea

space on 1 to 32 physical disks into a

single, larger logical drive.

WAN (Wide Area Network)

Pl
Re

A network spanning large areas such as

multiple cities, states, or countries.

well known addresses

Reserved service addresses for common
network processes.

524
 INDEX

ial
100BASE-FX, 121 Application Programming Interface Binary Synchronous Communications
100BASE-T4, 120 (API) See: BSC
100BASE-TX, 120 See: API binary values, 262
5–4–3 rule, 116-119 ARCnet, 102 bit mask, 285-286
80/20 Rule, 169-170 ARIN, 265-266 BNC, 77-78
802.3u Fast Ethernet specifications, ARP, 245-246, 294-295 boot PROMs, 150

er
120 troubleshooting, 307 BOOTP, 245-246, 247
ARPAnet, 225-226 bounded media, 70-83
A ATP, 210-211 BPDU, 177, 178-180, 181
access control Attachment Unit Interface (AUI) BRI, 322-323, 324
See: AUI bridge, 191-192

at
changing types, 342-343
share-level, 341-342 attenuation, 70-83, 135-136 bridge ports, 171-173
user-level, 342 AUI, 77-78 Bridge Protocol Data Unit (BPDU)

y
active headend, 95-96 AUI connector, 77-78 See: BPDU
active hubs, 138 authentication, 273-274 bridges, 7, 130-142, 169-184,

op
Active Monitor, 122
eM
Auto Negotiation (AUTONEG),
120
390
adapters, 142-153 bridge ID, 177
configuring, 146, 149 bridging loop, 174-177
B broadcast storm, 174-177

tC
installing, 151-152
troubleshooting, 152 backbone, 2-5 designating, 180
wireless, 145 backup domain controller failure, 181
address aggregation, 285-286 See: BDC fault tolerance, 174-177
No
Address Resolution Protocol (ARP) backup domain controllers, 18-20 relay entity, 173-174
c

See: ARP backups, 372 root, 177

address-to-name resolution, 205 data recovery, 410-411 selecting root bridges, 178-180
addresses large-capacity removable disks, source routing, 182-183
en

full, 189 406 transparent, 170-171

Do

operating, 189-190 media, 372, 405 bridging loop, 174-177

service, 189 methods, 377-379 British Naval Connector (BNC)
socket, 189 policies, 373 See: BNC
well known, 189 removable optical disks, 407 broadband transmissions, 94-96
small-capacity removable disks, brouters, 130-142, 192, 196, 197,
se

Alpha, 23-24
r

406 391
American Registry for Internet Num-
bers (ARIN) tape, 407 BSC, 326-327
fe

See: ARIN testing, 409 buffer space, 148

ea

amplifiers, 136-137 types, 373-374, 375-376 bus topology, 160-161, 164-166

analog connections, 321 bandwidth, 70-83, 139, 322, 324 buses
analog electronic devices, 92-93 base memory address, 148 EISA, 142-153
Pl

baseband transmissions, 94-96 ISA, 142-153

Re

analog signaling, 90-93

anti-virus engine, 413-414 Basic Rate Interface NuBus, 142-153
API, 54 See: BRI PC Card, 142-153
APNIC, 265-266 BDC, 18-20, 35 PCI, 142-153
AppleTalk, 27, 154-155, 204 BDR, 276 Vesa, 142-153
AppleTalk Session Protocol, 216-219 beacon frames, 125-126

Index 525
 INDEX

ial
C coaxial cable, 75-76, 77-78
coaxial connectors, 77-78
custom subnet mask, 279-285
cyclic redundancy check
cabling
codeset-and-data-structure protocols, See: CRC
cladding, 79 66-68 Cypress net, 225-226
coaxial, 75-76 Collision Detection, 113-114
fiber optic, 79 collisions, 101-102 D

er
grades, 79-81 complete trust, 36-37
IBM types, 72-73 DARPA, 224-229
concentrators, 130-142
jacket, 79 data encryption, 349-350
connection identifier, 204
plenum, 79-81 DES, 350
connection-oriented protocols,
shielded twisted pair, 71 digital signatures, 350
211-212

at
thicknet, 75-76, 115 private key, 350-351
connectionless protocols, 211-212
thinnet, 75-76, 115-116 public key, 350-351
connections
twisted pair, 71 RSA, 350
establishing, 220

y
unshielded twisted pair, 71 standards, 350
releasing, 220-221
data flow, 216-219

op
cabling connectors, 396
Caldera, 23-24
eM
transferring data, 220
connectors, 391
Data Link layer, 59-60, 61-62,
102-103, 145-146, 169-184,
CDPD, 84-85 AUI, 77-78 209193--212195
Cellular Digital Packet Data (CDPD) BNC, 392

tC
data rate, 70-83
See: CDPD coaxial, 77-78 Data Signaling
Channel Service Unit/Data Service D-type, 392 See: DS
Unit (CSU/DSU) DIX, 77-78 data-transmission media, 70-83
See: CSU/DSU fiber optics, 79, 392
No
DDPs, 276-277
channels, 84-85 IBM data, 393
c

DDS, 324-326
CHAP, 319 peripheral, 393-396 Debian, 23-24
checksums, 209-210 RJ-45, 73-74 DECnet LAT, 192
choke packets, 108
en

SMA, 79 default gateway, 199-200, 290-293

CIDR, 285-286
Do

ST, 79 configuring, 284-285

cladding, 79 terminator, 77-78 default gateway address, 239
Classless Inter-Domain Routing twisted pair, 73-74
(CIDR) default subnet mask, 279-285
ConsoleOne, 44-45 default subnetworks, 199-200
See: CIDR
contention, 101-102 Defense Advanced Research Projects
se

clients, 26-29
r

contention media-access method, Agency (DARPA)

LINUX, 28-29 154-155 See: DARPA
Macintosh, 27
fe

control program execution space, demand priority, 157-159

MS-DOS, 26
ea

148
demuxing, 140-142
Novell Client, 27 controllers, 155-156
deterministic media-access method,
peer-to-peer, 26-27 convergence, 268-270 156-157
protected mode, 27 count-to-infinity, 268-270, 272
Pl

device drivers, 419

Re

real mode, 26, 27 CRC, 105, 173-174, 193-195, device-to-device traffic, 209-212
Windows 3.x, 26 209-210
DHCP, 247, 248, 309
Windows 95, 27 crosstalk, 71
DHCP Server, 22-23
Windows 98, 27 CSMA/CA, 154-155
Dial-up IP, 225-226
Windows for Workgroups 3.x, CSMA/CD, 154-155
dial-up networking, 331-332
26 CSNET, 225-226

526
 INDEX

ial
dialog control backup domain controller, 18-20 file-transfer protocols, 66-68
full duplex, 216-219 multiple master, 35-36 filtering, 170-171
half duplex, 216-219 names, 237-238 finger, 254-255
simplex, 216-219 primary domain controller firewalls, 352
dialog control modes, 216-219 See: PDC floating-window flow control,
Digital Data Service single, 35 108

er
See: DDS single master, 35 flooding, 274
digital electronic devices, 92-93 trust relationships, 31, 32 flow control
Digital Service Windows NT, 34 guaranteed, 109
See: DS DOS, 26, 298-299 sliding window, 108
digital signaling, 90-93 downlinking, 86-87 static window, 108

at
direct memory access (DMA) DR, 276 stop and wait, 109
See: DMA drivers, 150-151 frame relays, 199-200

y
Direct Sequence Spread Spectrum DS, 324 frames, 59-60, 108, 145-146, 169-184
(DSSS) DSSS, 84-85 asynchronous, 103

op
See: DSSS
eM
dumb terminals, 13-14 beacon, 125-126
directory services, 29-47 Dynamic Host Configuration Protocol CRC, 104
disk duplexing, 18-20 (DHCP) isochronous, 103, 105-106
disk mirroring, 18-20, 355-356 See: DHCP parity bit, 104

tC
disk partitions, 368 dynamic routing, 198 persistently busy token, 123
disk striping, 18-20 start bits, 103
distance vector routing, 268-278 E synchronizing, 103
No
distributed processing, 13-14 electrostatic discharge synchronous, 103, 104
c
DIX connector, 77-78 See: ESD transmitting, 103
DLC, 196 email protocols, 66-68 Frequency-Division Multiplexing
DMA controller, 149 (FDM)
EMI, 71
en

DNA See: FDM

end-to-end communication, 209-212
Do

names, 237 FTP, 251-253, 254-255, 298, 308

end-to-end connection, 61-62
DNA Session Control, 216-219 commands, 298-299
end-to-end error control, 209-210
DNS, 231-232, 235, 254-255, troubleshooting, 309-313
end-to-end flow control, 210
290-293 FTP Server, 22-23
error detection
hierarchy, 235-237 full-duplex transmission, 217-218
se

parity bit, 104

r

zones, 242-243 full-neighbor state, 276-277

ESD, 386-387
DNS Client, 22-23
Ethernet, 64-65, 111-112, 154-155
fe

DNS Server, 22-23

external routers, 191-192 G
ea

documentation Gateway Service, 18-20

finding, 403-405
network, 402
F gateways, 7-8, 130-142
address, 7-8
failure, 181
Pl

vendor, 402-403
Re

default, 199-200
DOD model, 226, 231, 249 Fastwire, 10-12
format, 7-8
domain name space, 235-237 fault tolerance, 355, 360, 368
protocol, 7-8
Domain Name System (DNS) FDM, 140-142
Go Back N, 210
See: DNS fiber optic cable, 79
guaranteed flow control, 109
domains, 29-30, 36-37, 235-237 fiber optic connectors, 79
file-access protocols, 66-68

Index 527
 INDEX

ial
H Interior Gateway Protocol (IGP)
See: IGP
ISP, 321-322

half-duplex transmission, 218-219

HDLC, 326-327
internal routers, 191-192 J
Internet Control Message, 231-232 jabber, 437
headend, 95-96
Internet layer, 231-232
headers, 53-54, 100-101
Internet Network Information Center
L

er
Hello packet, 276 (InterNIC)
hierarchical proxy cache, 354-355 See: InterNIC LAN, 8-10, 62-64, 83, 136-137,
High-level Data Link Control 164-166, 169-184, 199-200,
Internet Protocol (IP)
See: HDLC 268218--270219, 284-285
See: IP
hops, 84-85, 195-196 cards, 142-153
internetwork connection devices,

at
host addresses, 238 drivers, 62-64
190-193
rules, 239-240 LANalyzer, 255-257
internetworks, 6
Host-to-Host layer protocols, 249 laser transmitter, 87-89
InterNIC, 260-268, 282-284

y
Hot Fix, 359-360 LAT, 196
interoperability, 7
latency, 121

op
Redirection Area, 360
HTTP, 251-253, 254-255
eM
interrupt request line (IRQ)
See: IRQ
layers, 52
hubs, 130-142, 157-159, 164-166, communication between, 53-54
interrupts, 147
390 Data Link, 59-60
exception, 147

tC
active, 137, 138, 162-163 headers, 53-54
hardware, 147
passive, 137, 138, 162-163 Host-to-Host, 227-229
software, 147
hybrid topology, 164-166 interaction among, 62-64
IP, 5-6, 205, 211-212, 231, 232,
238 , 247, 268-278, 279-285, Internet, 227-229, 231-232
No
I 294290--295293, 290, 291, LLC sublayer, 59-60
c

307294-295 MAC sublayer, 59-60

I/O addresses, 330-331
addressing, 260-268 Network, 58-59, 60-61, 66-68,
IANA, 266
IP Address Classes, 263-265 231
en

ICMP, 231-232, 233, 234

IP Next Generation (IPng) Network Access, 227-229
Do

IEE 802 specifications

See: IPng Physical, 59-60
Ethernet, 111-112
IP Version 6 Presentation, 55-56
IEEE 802 specifications, 110-126
See: IPv6 Process/Application, 227-229
Physical layer, 112
IPCONFIG, 304-305, 309 service access points, 53
IEEE 802.3 specifications, 112-113,
se

troubleshooting, 308 service data units, 54

r

116-119
IPv6, 267-268 Session, 56-57, 216-219
Ethernet, 114
IPX, 28-29, 211-212, 320 stacks, 53
fe

IEEE 802.5 specifications

IPX/IP Gateway, 22-23 transport, 57-58, 60-61, 66-68,
ea

Media Access Control, 122

IRQs, 147, 330-331 204-206, 216-219
Physical layer, 124-125
ISDN, 321, 323 Link State Request (LSR)
ifconfig
costs, 323 See: LSR
Pl

troubleshooting, 308
Re

ISDN connections Link State Update (LSU)

IGP, 268-278
BRI, 322-323 See: LSU
Integrated Services Digital Network
PRI, 322-323 link-state routing, 268-278
See: ISDN
ISM bands, 84-85 Linux, 23-24
interconnectivity, 7
Isochronous frame transmission, LLC, 59-60, 100-101, 102-103,
interface cards, 142-153 106-107, 109-110, 178-180
105-106

528
 INDEX

ial
connection services, 106-107 terrestrial microwave, 85-86 Directory tree, 38-40
error control, 109-110 unbounded, 70-83 Leaf object, 38-40
LLC eXchange Identification (XID) Media Independent Interface (MII), objects, 40-41
See: XID 120 partitions, 45
LMHOSTS, 205, 242 media-access control (MAC), 59-60 replicas, 45, 46
local area network (LAN) member servers, 18-20 [Root] object, 38-40

er
See: LAN mesh topology, 163 utilities, 44-45
LocalTalk LAN, 102 metropolitan area network (MAN) NDS Manager, 44-45
logical addresses, 188 See: MAN nearest active upstream neighbor
logical link control (LLC) MIB, 253-254 (NAUN), 125-126
See: LLC Microcom Networking Protocol NetBEUI, 28-29, 192, 196, 204,

at
loopback, 152-153 See: MNP 210-211, 320
LPD, 251-253 MILNET, 225-226 NetBIOS, 196, 292-293, 299-300
name resolution, 205

y
LSR, 276-277 MNP, 328
LSU, 278 modems, 133-135, 326 NetBIOS Name Server (NBNS)

op
eM
asynchronous, 326-327 See: NBNS
M configuring, 328-330 Netscape FastTrack Server, 22-23
Netscape Navigator, 22-23
installing, 328-330
MAC, 100-101, 102-103, 120,
ITU standards, 327 NETSTAT, 303-304

tC
125-126, 154, 246-247, 307
MNP standards, 328 troubleshooting, 308
addresses, 59-60, 102, 145-146,
170-171, 188 synchronous, 326-327 NetWare, 22-23
MAC header, 182-183 MSAU, 124-125 Admin, 23
bindery-based database, 22-23
No
MacLink Plus, 10-12 multi-point connections, 159-168
cross-platform communication,
c

mainframe-to-LAN connectivity, multicast packet, 273-274

121-122 multiple-session protocols, 66-68 25
MAN, 8-10 multiplexers, 139 NDS, 29-47
superuser, 23
en

Management Information Base (MIB), multiplexing, 140-142

Do

253-254 multipoint connection scheme, SUPERVISOR, 23

mandatory fields, 41-42 10-12 types of service, 22-23
media multiport repeater, 138-139 NetWare Administrator, 44-45
backup, 372 muxes, 140-142 NetWare Migration utility, 18-20
bounded, 70-83 NetWare Service Advertising Protocol
se

N
r

comparing types, 81-82, 89-90 (SAP)

contention, 154-155 See: SAP
narrow-band radio, 84-85
fe

data rate, 70-83 Network, 209-212

National Fire Protection Agency
ea

data transmission, 70-83 network accelerators, 354-355

(NFPA)
Ethernet types, 115 network interface card
See: NFPA
laser transmitter, 87-89 See: NIC
NAUN, 125-126
Network layer, 58-59, 61-62, 66-68,
Pl

media-access, 154-155
Re

NBP, 210-211
100-101, 102-103, 110-126,
method, 154-155 NBTSTAT, 299-300 193188--195190, 209-212, 231,
microwave, 83-90 troubleshooting, 308 255-257
radio, 84-85 NDS, 22-23, 29-47 network operating system
satellite microwave, 86-87 container object, 38-40 See: NOS
storage, 409 database components, 38 network prefixes, 285-286

Index 529
 INDEX

ial
network services, 18-25 physical conditions, 385 OSI model, 50-65, 110-126, 133-135,
Network Terminal Interface protocol stack, 5-6 145-146, 204, 211-212, 226,
See: NT1 protocol suite, 5-6 354354--355355
network transparency, 64-65 security, 340 layers of, 50-65
network-management protocols, software, 397 OSPF, 268-278
66-68 spatial conditions, 388-389 Hello packet, 276

er
networking strategies temperature precautions, 385-386 route discovery, 276
centralized, 12-16 Token Ring, 64-65, 72-73, 121-122, synchronization, 276-277
client-server, 12-16 124-125, 162-163 overview, 52
hierarchical, 12-16 topologies, 10-12
host-to-terminal, 12-16 twisted pair, 73-74 P

at
legacy, 12-16 next hop addressing, 273-274 packets, 157-159
node-connectivity, 12-16 NFPA, 79-81 PAP, 319

y
peer-to-peer, 12-16 NFS, 251-253 parity bits, 104
primary-to-secondary, 12-16 NIC, 64-65, 100-101, 102, 131, partitions, 34

op
networks
eM
132, 133-135, 142-153, 389
NNTP, 254-255
pass-through authentication, 29-30
802.5, 121-122 passive headend, 95-96
ARCNet, 64-65 nodes, 2-10, 13, 14, 209-212 passive hubs, 138
backbone, 2-5 NOS, 18-25, 26, 382

tC
passwords, 348, 383
building, 83 Novell Directory Services (NDS) patch cables, 390
classes, 239-240 See: NDS patch panels, 389
cleanliness precautions, 387 NSFnet, 225-226 patches, 417, 419-421
NSLOOKUP, 307
No
client-server, 13-14, 216-219 applying, 418
NT1, 323
c
client/server, 56-57 locating, 418
clients, 26 NTDS, 29-30 testing, 418-419
components, 2-5, 389 null modem, 133-135 PBX, 322-323
en

configuring clients (TCP/IP), NWLink, 210-211 PDC, 18-20, 35

Do

290290--293293 PDU, 102-103

defined, 2-10 O peers, 14
documentation, 402 objects peripherals, 391
domains, 18-20 Container, 42-43 Physical layer, 59-60, 100-101,
electrostatic discharge precautions, default NDS, 44 135-136, 169-184
se
r

386-387 PING, 306

Leaf, 42-43, 44
environmental issues, 385 troubleshooting, 308
NDS, 40-41
fe

hardware, 397 Plain Old Telephone Service

properties, 41-42
ea

hierarchical, 13-14, 56-57, 216-219 See: POTS

[Root], 42
installing, 382 plenum cable, 79-81
third-party APIs, 41-42
layers, 51, 52 Point-To-Point
types, 42
Pl

legacy, 13-14
Re

values, 41-42 See: PPP

mixed, 15 point-to-point connections, 10-12,
Open Shortest Path First
Network Ring, 183-184 159-168
See: OSPF
NTDS, 29-30 Point-to-Point Tunneling Protocol
Oracle, 22-23
peer-to-peer, 13-14, 21, 56-57, See: PPTP
216-219 poison reverse, 268-278
personal items, 387

530
 INDEX

ial
polling, 101-102, 155-156 network management, 66-68 remote-action protocols, 66-68
POP3, 251-253, 254-255 reliability, 66 repeater count, 136-137
port states, 171-173 remote action, 66-68 repeaters, 121, 130-142, 169-184,
ports, 254-255 routing, 66, 196 390
POSIX, 23-24 spanning tree, 177, 181-182 multiport, 137, 138-139
POTS, 321 task-to-task, 66-68 signal regenerating, 136-137

er
PPP, 318, 319, 322 terminal emulation, 66-68 source routing, 169-184
PPTP, 320 unroutable, 196 source-routing transparent, 169-184
prefix length, 285-286 proxy cache servers, 354-355 transparent, 169-184
Presentation layer, 55-56 proxy servers, 352 replicas
PRI, 322-323, 324 PSTN, 321 master, 46

at
Primary Rate Interface Public Switched Telephone Network read-only, 46
See: PRI See: PSTN read/write, 46

y
print servers, 391 Pyramid, 23-24 subordinate references, 46
priority control, 123 request for comments

op
private addressing, 266 R
eM See: RFC
Private Branch Exchange resolvers, 243-245
radio, 84-85
See: PBX Reverse Address Resolution Protocol
cellular phone service, 84-85
(RARP)

tC
private network addressing, 239 narrow band, 84-85
Process/Application layer protocols, See: RARP
spread spectrum, 84-85
251-253 ring topology, 161-162
RAID, 363
programmable Read Only Memory RIP, 273-274
mirroring (Level 1), 364-365
(PROM) RIP II, 273-274
No
striping (Level 0), 363-364
See: boot PROMs RIPE, 265-266
c

striping by block with parity on a

promiscuous mode, 437 RJ-45 connectors, 73-74
single drive (Level 4), 366
propagation time delay, 136-137 crimping, 74-75
striping with error correction (Level
en

properties 2), 365 rotation methods, 374-375

Do

mandatory fields, 41-42 striping with parity spread across Tower of Hanoi, 374-375
protected mode, 26-27 multiple drives (Level 5), ROUTE command, 302-303
Protocol (ICMP) 366366--367367 route database, 276-277
See: ICMP stripping with parity on a single route discovery, 183-184
Protocol Data Unit (PDU) drive (Level 3), 365 route selection, 197
se
r

See: PDU RARP, 245-246, 247, 295-296 comparison of type, 198-199

protocol stack, 5-6 read-after-write verification, 360 dynamic, 198
fe

protocols, 5-6 receiver, 2-5 static, 198

ea

codeset and data structure, 66-68 Red Hat, 23-24 route tables, 195-196
connection, 66 Redundant Array of Inexpensive Disks router discovery, 234
connection oriented, 211-212 See: RAID routers, 7, 130-142, 191, 193-200,
Pl

connectionless, 211-212 relay agents, 248 390

Re

email, 66-68 relay entity, 173-174 external, 191-192

families, 66-68 release notes, 402-403 internal, 191-192
file access, 66-68 remote access Routing Internet Protocol (RIP)
file transfer, 66-68 PPP, 318 See: RIP
multiple session, 66-68 SLIP, 318 routing protocols, 196
remote-access connections, 331-332

Index 531
 INDEX

ial
S SMA connectors, 79
SMTP, 251-253, 254-255, 308
Supernetting, 285-286
Suse, 23-24
SAP, 216-219
SNA, 155-156, 196 switches, 90-92, 130-142, 149
scanning, 414-415
Sniffer, 255-257 switching hub and routers, 196,
heuristic, 414-415 197
SNMP, 251-253, 254-255
local, 414-415 Synchronous Data Link Control
agents, 253-254

er
on access, 414-415 See: SDLC
managers, 253-254
on demand, 414-415 synchronous frame transmissions,
socket addresses, 189
online, 414-415 104
sockets, 254-255
SDLC, 326-327 SYSCON, 44-45
SOPs, 383-384
sector sparing, 359 System Fault Tolerance
source-routing bridges, 182-183

at
Windows NT, 359 See: fault tolerance
spanning tree protocol, 177
security Systems Network Architecture (SNA)
Sparc, 23-24
levels, 340-341 See: SNA

y
split horizon, 268-278
password, 347-348
spread-spectrum radio, 84-85
T

op
shared-level, 346
UNIX, 345-346
eM
SPX, 204, 210-211
ST connectors, 79 TA, 323
user-level, 346
stacks, 53 task-to-task protocols, 66-68
security models, 339

tC
Standard Operating Procedures TCA, 181
segments, 130-142, 181
See: SOPs TCN, 181
sequencing, 208-209
standards TCP, 5-6, 204, 210-211, 249, 308
synchronizing, 208-209
de facto, 162-163 TCP/IP, 27, 28-29, 199-200, 205,
Selectively Repeat, 210
No
de jure, 162-163 224-229 , 231, 238, 246-247,
Sequent, 23-24 268251--270253, 279-285, 291,
c

star topology, 162-163

Serial Line Internet Protocol 296-297, 320, 383
start bits, 103
See: SLIP configuration, 248
static routing, 198
service access points, 53 displaying current client
en

Statistical time-division multiplexing

configuration, 304-305
Do

service address, 189 (STDM)

service data units, 54 suite utilities, 294
See: STDM
service-provider initiated method, TDM, 140-142
STDM, 140-142
205 TE, 323
stop-and-wait flow control, 109
service-requester initiated method, Telnet, 251-253, 254-255, 296-297
store-and-forward, 173-174
se

205
r

troubleshooting, 307
STP, 73-74
Session layer, 56-57, 61-62, 215-221 terminal adapter
stripe sets, 370-371
SFT
fe

See: TA
striping, 363-364
See: fault tolerance
ea

terminal equipment
structural influences, 29-30
shared resources See: TE
subdomains, 235-237
accessing, 345 terminal-emulation protocols, 66-68
subnet mask, 199-200, 273-274,
shielded twisted-pair cabling, 71-72 terminator, 77-78
Pl

290
Re

signal regenerating repeater, 136-137 TFTP, 251-253, 254-255

subnet masks, 280-281
signaling thicknet cabling, 75-76
subnets, 191-192, 279-285
transmissions, 94-96 thinnet cabling, 75-76
subnetworks
simplex transmission, 216-217 throughput, 70-83, 169-184
default, 199-200
sliding-window flow control, 108 Time-Division Multiplexing (TDM)
subscription services, 403-405
SLIP, 318, 319, 322 See: TDM

532
 INDEX

ial
token passing, 101-102 unacknowledged connectionless UDP, 204, 211-212, 249, 250-251
Token Rign network, 183-184 service, 209-212 unbounded media, 70-83
tokens, 156-157 Transport layer protocols, 210-211 uninterruptible power supply
passing, 156-157 traps, 253-254 See: UPS
topologies, 10-12, 159-168 failure to authenticate, 253-254 Universal Asynchronous Receiver/
active, 161-162 link up or down, 253-254 Transmitter

er
bus, 10-12, 160-161, 164-166 warm/cold boot, 253-254 See: UART
cellular, 10-12 trojan horse, 412 UNIX, 15, 18-25, 205, 225-226,
dual counter, 161-162 troubleshooting 298-299, 304-305, 307, 318, 382
dual ring, 161-162 cable testers, 436 cross-platform communication,
Collect, Isolate, and Correct model, 25
hybrid, 10-12, 164-166

at
431-435 portability, 23-24
linear bus, 164-166
crossover cables, 435-436 root, 25
logical, 10-12, 159-168
hardware loopback plugs, 436 superuser, 25

y
mesh, 10-12, 163
hardware tools, 435 unroutable protocols, 196
physical, 10-12, 159-168

op
ring, 10-12, 161-162
eM
manufacturer resources, 440-442
methodology, 424
unshielded twisted-pair cabling,
71-72
rotating ring, 161-162
models, 426-427 uplinking, 86-87
star, 10-12, 162-163
NetWare log files, 438-439 UPS, 357

tC
Topology Change Acknowledgment
Network+ model, 427-429 Linux, 359
(TCA)
Novell model, 429-431 monitoring, 357
See: TCA
performance monitors, 437-438 NetWare, 358-359
Topology Change Notification (TCN)
protocol analyzers, 437 Windows NT, 358
No
See: TCN
software tools, 436-437 USP, 391
c
Tower of Hanoi, 374-375
tone generator, 436 UTP, 73-74, 116
TRACERT, 301-302
troubleshooting, 308 tone locator, 436
V
en

trailers, 100-101 tools, 435

Do

transaction identifier, 204 Windows NT log files, 439-440 vendors

Transaction Tracking System trusts documentation, 402-403
See: TTS complete, 36-37 video cards, 148
transceiver, 2-5, 77-78, 131 establishing, 33-34 virtual private network
Transmission Control Protocol (TCP) guidelines, 32-33 See: VPN
se
r

See: TCP one-way, 32, 35 virus definition file, 413-414

transmission media adapters, 132-133 single domain, 35 virus protection, 412
fe

transmission media connectors, two-way, 32, 35-36 virus signature, 413

ea

133 Windows NT models, 34 viruses

transmitter, 2-5 TTS, 361-363 hoaxes, 413
transparent bridge, 170-171 twisted-pair cabling, 71-72 prevention, 413
Pl

Transport layer, 57-58, 61-62, twisted-pair connectors, 73-74 protection, 413-414

Re

66-68, 204-206, 216-219 twisted-pair networks, 73-74 types, 412

acknowledged connectionless two-way state, 276-277 volume sets, 368-370
service, 209-212 VPN, 320
connection services, 209-212
connection-oriented service, 209-212
U
UART, 321
segmenting process, 207-209

Index 533
 INDEX

ial
W
WAN, 8-10, 18-20, 164-166, 196,
231-232, 284-285, 324-326,
354354--355355
Web proxy cache servers
See: proxy cache servers

er
well-known addresses, 189
white papers, 402-403
wide area network (WAN)
See: WAN

at
Windows NT, 18-20
administrative utilities, 30-31
Administrator, 21

y
BackOffice, 18-20
cross-platform communication,

op
25
eM
Directory Services, 29-47
domain models, 34

tC
editions, 18-20
fault tolerance, 18-20
global groups, 29-30
local groups, 29-30
No
replication, 34
c

superuser, 21
trust relationships, 31
trusts, 32-33
en

workgroup model, 21
Do

Windows NT Directory Services

(NTDS)
See: NTDS
WINIPCFG, 304-305, 309
troubleshooting, 308
se
r

WINS, 205, 290-293

process description, 205-206
fe

wireless adapters, 145

ea

workgroup model, 21

X
Pl
Re

X25Net, 225-226
XID, 183-184
XNS, 204

534