OSPF Characteristics

OSPF is a Link State Routing Table. Each router will have complete topology
information.
OSPF breaks the network down into logical groups of routers known as Areas. In
this video, we will focus on a Single Area.
Just like EIGRP, OSPF also uses a Hello protocol to discover neighbors and form
Neighbor relationships. OSPF uses 224.0.0.5 & 224.0.0.6 Multicast addresses to
communicate to its neighbors.
As each router has complete topology information, it can quickly adapt to alternate
paths. If the primary route in the routing table is lost, the router will recalculate an
alternate path (if it exists) immediately without consulting a neighbor.
OSPF supports Equal Cost Load Balancing.
OSPF Characteristics
 OSPF supports the validation of neighbors by using Authentication using Clear Text and MD5
based Keys. Authentication makes sure that the routes received by the router are from
Trusted sources / Routers.

 OSPF also supports the Summarization of routes. It allows you the ability to send a summary
route to represent a group of routes allowing you to cut down on the size of the routing table.

 OSPF advertises the network along with it’s subnet mask making it a Classless Routing
Protocol. It supports features like VLSM.

 OSPF sends information about all the directly connected Links to its neighbor. The updates
are acknowledged. Just like EIGRP, OSPF does not send the Full routing table. The Hello
Packets are used as keepalives to maintain the neighbor relationships.
OSPF Characteristics - Timers
Hello Timer: Once the OSPF neighbors establish a
neighbor relationship and exchange Network information,
they send Hello packets to let the neighbor know that it is
still up and the routes are still valid. The default OSPF
Hello time on a Ethernet Link is 10 seconds.
Dead Timer: If a router does not receive a hello for 40
seconds from a neighbor, it will consider the neighbor
down and delete all the routes from it.
OSPF Characteristics – Metric
OSPF uses a Cost Metric.
By default, the cost is calculated based on the logical interface
bandwidth.
The metric calculation formula for OSPF is :

Cost = [Reference Bandwidth/Interface Bandwidth(sum)]

The default Reference Bandwidth is 108 which is 100,000,000. It is calculated in

bps. You do have the ability to change the reference bandwidth to accommodate
interface with high speed using the “Auto-cost reference-bandwidth” command.
OSPF – Router ID
In OSPF, each router is identified internally within OSPF using an Identification
number which looks like an IP Address. This is known as a Router-ID.
Although it looks like an IP address, it is not an IP Address and does not need to
be advertised. You could actually assign a Router an ID that is an invalid IP
Address but a valid Router-ID like 0.0.0.1.
Router ID is required for each OSPF router.
It can be set either of the 3 ways and in that order.
1. Manually under the Router ospf process using the router-id command.
2. If the command is missing under the process, OSPF tries to pick the router-id based on the
Highest IP address on a Loopback.
3. If you don't have a loopback interface, it will use the highest IP address on any Physical
Interface.
OSPF – Ethernet Media
OSPF runs differently on different types of Interface.
On an Ethernet segment, it treats the interface as a Broadcast Multi-Access Segment. It has
the ability to have multiple neighbors.
Rather than establishing a neighbor relationship with multiple neighbors, OSPF elects a
Router on the segment as the Designated Router [DR] for the Segment and establishes a
neighbor relationship only with that Router.
For redundancy, every segment also has a Backup Designated Router [BDR]. All routers also
establish a neighbor relationship with the BDR.
Every Multi-Access Segment will have its own DR/BDR. The main purpose is to cut down on
the number of relationships that a router needs to maintain on a Multi-Access.
OSPF detects a Multi-Access segment by looking at the Encapsulation of the Interface. When
it sees ARPA as the encapsulation, it knows it is Ethernet, which has a potential of being Multi
Access and hence will go into a DR/BDR process.
OSPF – Serial Media
OSPF detects a Point-to-Point Interface by looking at the
Encapsulation of the Interface. When it sees either
HDLC or PPP as the encapsulation, it knows it is a Point-
to-point segment.
A P2P segment will only have 1 neighbor hence does not
require a DR or BDR.
There is no DR or BDR on a Point – To – Point Segment.
OSPF – Loopback Interface
OSPF recognizes a Loopback interface as a Dead End Interface
and hence advertises it as a Host.

The Neighboring routers will receive the Network as a Host

Route [/32 route].

You can have it advertise the normal interface mask by

manually changing the network type of the interface from the
default Loopback to Point-to-point.
OSPF LSAs
OSPF uses Link State Advertisements [LSA] to build the topology of the
network.

LSAs are the building blocks of the OSPF Link State Database [LSDB].
The LSDB is similar to your EIGRP topology table.

Although there are multiple types of LSAs in OSPF, only 2 LSAs are
used for network/routes within a Single Area.

In this video we are going to look at the first 2 LSAs

OSPF – LSA Type 1 – Router LSA
Every Router will advertise it's directly connected links to
the routers on the local Segment.

Those Routers in turn will propagate the information to

other routers in the area without changing it.

This information remains unchanged within the Area

and is not propagated out of the area.
OSPF – LSA Type 2 – Network LSA
This LSA is advertised by the DR of a Multi-Access Segment.

It will advertise the Multi-access segment and specify the

attached Routers.

It is used to validate information from LSA Type 1 regarding any

multi-access segments.

It also specifies the Subnet Mask for the Multi-Access Segment.
Basic OSPF Configuration
All Routing Protocols are enabled by using the “Router” command.

In order to enable OSPF, you use the “Router OSPF” command followed by a process
ID.

The Process is locally significant. It does not need to match the neighboring router.

To enable OSPF, you type

R1(config)#Router ospf 1
R1(config-router)#
Basic OSPF Configuration
OSPF Routers requires a unique identifier. This ID is known as the
Router ID.
It can be set either of the 3 ways and in that order.
1. Manually under the Router ospf process using the “router-id” command.
2. If the command is missing under the process, OSPF tries to pick the router-id
based on the Highest IP address on a Loopback.
3. If you don't have a loopback interface, it will use the highest IP address on any
Physical Interface.

R1(config)#Router ospf 1
R1(config-router)#router-id 0.0.0.1
Network Command
Just like the other Routing protocols, the “Network” Command is used by the Routing
Protocols on the specified Interface.
By Enabling the routing protocol under the interface, it will take the following 2 actions:
• 1. It will allow the router to send and receive OSPF updates on that interface.
• 2. It will send/advertise this interface network on other OSPF enabled interfaces.
In OSPF, the network command allows you the ability to enable OSPF on a Major network or
subnets. The subnets can be specified using a Inverse Mask.
The Area that the interface belongs to is also specified in the Network Command.
You enable OSPF on all the directly connected interfaces.
R1(config)#Router OSPF 1
R1(config-router)#Network 10.1.1.0 0.0.0.255 area 0
R1(config-router)#Network 11.0.0.0 0.255.255.255 area 0
R1(config-router)#Network 192.1.13.0 0.0.0.255 area 0
Configuring OSPF on a Point – To
– Point Serial Interface
OSPF detects a Point-to-Point Interface by looking at the
Encapsulation of the Interface. When it sees either HDLC or PPP
as the encapsulation, it knows it is a Point-to-point segment.

A P2P segment will only have 1 neighbor hence does not require
a DR or BDR.

There is no DR or BDR on a Point – To – Point Segment.

OSPF on a Ethernet Segment
When you enable OSPF on a Ethernet Interface, it knows it is a Multi-Access Segment.
It will look for a DR on the segment. All neighbors will form a Neighbor relationship with the DR &
BDR on the segment.
Each Multi-access segment will have its own DR/BDR.
If it is the first router on the Interface, it will wait to conduct a DR/BDR election. The wait time by
default is 40 seconds.
The DR/BDR election criteria is as follows:
1. Highest Interface Priority [OSPF Priority]. The default OSPF priority is 1. Higher the Better.
2. Router-id is the Tie Breaker. As the Router ID has to be unique, the highest Router ID device becomes the DR.
There is no Pre-emption in the DR/BDR election.
To set the Interface Priority for OSPF, use the following Interface specific command:

Interface E 0/0
ip ospf priority 10
Verifications
Type Show IP OSPF Neighbors on R5. You should see 2 Neighbors listed.

Type Show IP route OSPF on R2. You should see the Loopback Interfaces of R4
& R5. They will be listed as a Host router “/32”. This is due to the fact that
OSPF treats a Loopback as a Host and advertises it with a /32 mask.

Type Show IP OSPF Interface S 1/0 on R4. The Link will have a Network Type
of Loopback.

Type Show IP OSPF Interface S 1/0 on R2. The Link will have a Network Type
of Point-to-point.
OSPF Characteristics – Metric
OSPF uses a Cost Metric.
By default, the cost is calculated based on the logical interface
bandwidth.
The metric calculation formula for OSPF is :

Cost = [Reference Bandwidth/Interface Bandwidth(sum)]

The default Reference Bandwidth is 108 which is 100,000,000. It is calculated in

bps. You do have the ability to change the reference bandwidth to accommodate
interface with high speed using the “Auto-cost reference-bandwidth” command.
OSPF Multi-Area Characteristics
 Single-Area OSPF design puts all routers into a Single OSPF area. This
results in many LSAs being processed on every router.
 Each router maintains detailed information about the links and routers
within the area.
 Each link will have a LSA advertised by multiple router.
 For example, If you have 3 routers connected to a segment
192.168.100.0/24, all 3 routers will advertise that they are connected
to the segment. In addition to that, the DR will advertise LSA Type 2 for
the same segment. The LSDB will have 4 entries for the same segment.
OSPF Multi-Area Characteristics
In a Multi-Area design, network is divided to limit the propagation of LSAs.
The router will maintain general information about links in other areas. The
router that connects 2 area’s to each other is known as the Area Border
Router (ABR). It will advertise the network only once to routers in other areas.
Summarization is not supported within the Area as each router needs to have
complete information of the Area. Summarization is allowed on the ABR.
Another advantage of Multi-area is to limit the scope of LSA Flooding. When a
link state changes, it will only get propagated within the area, unless it
specifically affects the entire design.
OSPF Multi-Area Structure
OSPF supports a hierarchical network structure that has a two-layer
hierarchy. The two-layers are:
• Backbone / Transit Area: The primary function of this OSPF area is to
interconnect other areas to each other. The OSPF Hierarchical area structure
requires that all area connect directly to the backbone area. Generally, end
users are not found within a backbone area. The backbone area has to have
the Area ID as 0. The backbone area and Area 0 are synonymous.
• Normal / Non-Backbone Area: Normal areas are usually setup to functional
or geographical groupings. By default, a normal area does not allow traffic
from another area to use its links to reach other areas. Generally, the end
users and resources like servers are located in normal areas.
OSPF Router Types
Backbone Routers: Routers that have a link in Area 0 are known as
Backbone routers. OSPF hierarchical structure defines Area 0 as the core. The
function of the backbone area is to connect other areas to each other. This is
implemented by routers within Area 0.

Area Border Routers: Routers in Backbone area that connect to other areas
are known as Area Border Routers (ABR). As the ABR belongs to multiple
areas, it will maintain complete information about the topologies in both areas.
It separates the LSA Flooding zones by sending general (Summary) information
about one area to another. If required, Route Summarization is performed on
the ABR.
OSPF Router Types
Internal Routers: Routers that have all links within the same area is known
as a Internal Routers. Backbone routers which have all their links in Area 0
can also be classified as Internal Backbone Routers.

Autonomous System Boundary Routers (ASBR): This router connects OSPF

to a Non-OSPF Routing domain. For example, if some part of your network is
running EIGRP and other part of your network is running OSPF, the EIGRP
routes will not get injected into OSPF and vice versa by default. You do have
the ability to configure Route Redistribution to inject routes from 2 different
domains into each other. The Redistributing Router in OSPF is known as the
ASBR.
OSPF LSA Types
LSAs are the building blocks of the OSPF LSDB.

LSAs are Database records which allow your router

to build the entire topology.

We will discuss the most commonly used LSAs in the

following slides.
OSPF – LSA Type 1 – Router LSA
Every Router will advertise it's directly connected links to
the routers on the local Segment.

Those Routers in turn will propagate the information to

other routers in the area without changing it.

This information remains unchanged within the Area

and is not propagated out of the area.
OSPF – LSA Type 2 – Network LSA
This LSA is advertised by the DR of a Multi-Access Segment.

It will advertise the Multi-access segment and specify the

attached Routers.

It is used to validate information from LSA Type 1 regarding any

multi-access segments.

It also specifies the Subnet Mask for the Multi-Access Segment.
OSPF – LSA Type 3 – Summary LSA
An ABR takes the detailed information that it learns from one
area and summarizes it for another area.

For example, if there are 3 LSA Type 1’s advertised by 3 routers

connected to a single segment (192.168.100.0/24), the ABR
sends 1 LSA to the other area informing the routers that if they
need to reach the network, they can go thru the ABR to get to it.

This is done automatically by the ABR.

OSPF – LSA Type 5 – External LSA
Routes that are redistributed into OSPF from other
Routing Domains like EIGRP are advertised as LSA
Type 5.

Route redistribution is done by ASBRs. The External

LSA Type 5 is advertised by the ASBR.