Veritas Netbackup
Veritas Netbackup
1. Introduction & Netbackup Architecture. 2. Installation of Master / Media / Clients software, type of backups and archives 3. Demons and services 4. Backup Hard ware devices (Library and Tapes) 5. Configuration of Policies 6. Configuration of Storage units / Volume pools 7. Netbackup Error Codes and Trouble shooting 8. Important Netbackup commands 9. Restoration / Recovery 10. Health Check & Intializing scheduled backup & Daily job monitoring
Information Storage
Importance of information Data, information and storage Categories of data Storage architectures and their evolution
What is Data
Collection of raw facts from which conclusions may be drawn
Data is converted into more convenient form i.e. Digital Video Data Increase in data processing capabilities Lower cost of digital Photo storage Affordable and faster communication technology
Book
Digital Data
Categories of Data
Data can be categorized as either structured or unstructured data Over 80% of enterprise information is unstructured
Contracts E-Mail Attachments X-Rays PDFs
Check
Unstructured (80%)
Instant Messages Documents
Manuals
Images
Forms
Structured (20%)
Rows and Columns
Information
What do individuals/businesses do with the data they collect?
Centralized information storage and processing Network Network
They turn it into information Information is the intelligence and knowledge derived from data
Wired
Wireless
Wireless
Wired
Uploading information
Accessing information
Creators of information
Users of Information
Storage
Data created by individuals/businesses must be stored for further processing Type of storage used is based on the type of data and the rate at which it is created and used Examples: Individuals: Digital camera, Cell phone, DVDs, Hard disk Businesses: Hard disk, external disk arrays, tape library Storage model: An evolution Centralized: mainframe computers Decentralized: Client server model Centralized: Storage Networking
Introduction to Netbackup
Veritas Product Symantec took over veritas 70 to 80% of companies world wide use Versions->5.1 to 6.5.37,7.01 GUI and CLI
Netbackup Client
During the backup or archive,the client system generates the backup as a data stream and sends it across the network to the media server. Therefore, our target is the client whose data needs to be backed up or restored MS SQL Agent comes free with Netbackup client
Media Server
To increase performance by providing additional processing ability and better distribution of workload (Load balancing). Note: Tape libraries are always attached to the media server only
Master Server
Master Server is netbackups central brain, providing centralised operations and management to any media server and clients
ARCHIVE BIT
The archive bit is a binary bit. As such, it is either 1 or 0, also frequently called set (1) and clear or reset (0). It represents whether the file has (1) or has not (0) changed since the last backup.
Types of Backup
Full (Archive bit is reset) Differential Incremental (Changes since the latest backup, archive bit is reset) [Incremental] Cumulative Incremental(Changes since the last full backup) Archive Backup (Delete data from source after backup leaving behind the metadata ,eg:Oracle archive log backup) User Initiated Backup (Backup window is 24x7)
Vault
Vault job initiates duplicate backup jobs. One for onsite and another for offsite Ejects the filled tapes from the tape library Automatically sends email to Iron mount and backup admins stating the scratch set tape details and tapes which need to be reloaded License is very costly.
NetBackup Catalog
NetBackup catalogs are internal databases that contain information about NetBackup backups and configuration. Backup information includes records of the files that have been backed up and the media on which the files are stored. The catalogs also contain information about the media and storage devices. NetBackup requires the catalog information in order to restore backups so it is extremely important to configure a catalog backup before using NetBackup for regular client backups, and to schedule the catalog backups to occur on a regular basis thereafter. Without regular catalog backups, you risk losing your regular backups if there is a problem with the disk that contains the catalogs.
Netbackup Management
Host Properties Master Server Media Server Clients ->Double click to check network connectivity
NOTE
Media ID:- is a 6 digit alphanumeric & usually a media has a barcode. Media>Robots (RC)>Robot Inventory>Update Volume config>Start
Frozen,Suspend,Import,Duplicate
Frozen tape:-Media Manager will auto freeze a media if the backup cant use it after 3 attempts. Incase of media errors 83,84,86,we freeze the media. Unfreeze later. Suspend:-It means that Netbackup won't write any new data to that tape, until all data on it expires. It will then be available for re-use. Expire:-Data on media will be removed but the image (index) will remain until the tape is overwritten Duplicate:-Medias sent offsite.Backup done using either vault/duplicate backup option. Import:-catalog db(index/image) from tape when server crashes and needs restoration
Volume Pool
Definition:-Volume Pool is a set of volumes from which netbackup can draw or take volumes when performing backup Group of tapes -Volume/Cartridge/media/tape A volume pool identifies a logical set of volumes by type of usage. Associating volumes with a volume pool protects them from access by unauthorized users, groups, or applications. With the exception of the volume pools that are automatically created by NetBackup, you must create a volume pool before you can add volumes to a volume pool. The following volume pools are automatically created (see Volume Pools List): NetBackup, created by Media Manager, for NetBackup use. DataStore, created by NetBackup, for DataStore use. CatalogBackup, created by NetBackup, for NetBackup catalog backups. During initial configuration, it is easiest to create all of your volume pools first. Then as you add volumes, you can assign them to volume pools. It is also possible to configure a scratch pool from which Media Manager can transfer volumes, when a volume pool has no volumes available. If the volume pool will contain WORM media, see Using WORM Volume Pools to Manage WORM Media for more information.
Volume Groups
Volume Groups is a physical location for a media. It manages a group of tapes for purpose of admin actions that allows movement of tapes from slots to drive and vice versa. Rules for Assigning Volume Groups The following are the rules for assigning volume groups: All volumes in a group must be the same media type. However, a media type and its corresponding cleaning media type are allowed in the same volume group (for example, DLT and DLT_CLN). All volumes in a robotic library must belong to a volume group. You cannot add volumes to a robotic library without specifying a group or having Media Manager generate a name for the group. The only way to clear a volume group name is to move the volume to standalone and not specify a volume group. More than one volume group can share the same location. For example, a robotic library can contain volumes from more than one volume group and you can have more than one standalone volume group. All volumes in a group must be in the same robotic library or be standalone. That is, Media Manager will not let you add a group (or part of a group) to a robotic library, if it already exists in another robotic library.
What is a Tape?
Tape began as the primary storage medium for computers in the 1950s. At that time, large mainframe computers dominated the information systems industry. Data storage and backup for these mainframes was on expensive, reel-to-reel tape systems using half-inch magnetic tape. Data was recorded in parallel tracks using a linear serpentine recording technology in which the data tracks were written in alternating bands from the beginning of the tape to the end and back again. Throughout the 60s and 70s these open-reel systems evolved by increasing the number of tracks, increasing bit density, and increasing tape speed.
SDLT
Super DLT tape technology is the next generation DLT tape technology platform, the current industry standard for data storage Super DLT tape technology is based on Laser Guided Magnetic Recording (LGMR) technology, which uses a unique combination of the best optical and magnetic technologies to dramatically increase the number of recording tracks on the data-bearing surface of the media The storage capacity of a SDLT-220 drive is 110GB Native and 220GB Compressed. The storage capacity of a SDLT-320 drive is 160GB Native and 320GB Compressed.
LTO Tape
Speed and capacity is expected to double with each generation of the LTO format. The LTO format Generation 1 allows for up to 100GB capacity (uncompressed) in a single cartridge, while LTO format Generation 2 doubles the capacity up to 200GB (uncompressed) using Generation 2 data cartridges. The data transfer peak performance speed doubles from up to 20MB/second native in the LTO format Generation 1 up to 40MB/second native in Generation 2. Users will benefit from a greater than 30 percent increase in the number of tracks and average tape speed, as well as an improved recording method. The LTO format Generation 2 provides users with outstanding data and investment protection because of its capability to read-and-write Generation 1 cartridges in the LTO Generation 1 format. LTO format come in 2 flavours : 1) Single reel catridge optimized for greater storage capacity called ULTRIUM and 2) Dual reel cartridge designed for faster access called ACCELIS
LTO Tape
The LTO compatibility is defined with two concepts demonstrating investment protection: 1) An LTO drive is expected to read data from a cartridge in its own generation and at least the two prior generations. 2) An LTO drive is expected to write data to a cartridge in its own generation and to a cartridge from the immediate prior generation in the prior generation format
LTO-5
The current draft has the following specifications:[14][15] Capacity of 1.5 TB (1500 GB) Uncompressed data transfer rate of 140 MB/s Specifications announced January 19, 2010 The first LTO-5 drives appeared on the market in Q2, 2010 [16]
EXABYTE - Mammoth
Exabyte introduced the 8mm helical scan tape drive in 1985. At that time, the 8mm drive mechanical subassembly was designed and manufactured by Sony. Exabyte supplied the electronics, firmware, and cosmetics for the final product. Two generations of full high and one generation of half high tape drives were developed under the Sony/Exabytepartnership. Exabyte chose to develop and manufacture their own scanner for the Mammoth product and terminated their relationship with Sony for any future drivetechnology. Exabyte continues to rely on Sony for its Mammoth Advanced Metal Evaporative (AME) media. Exabyte also uses mechanical sub-assemblies from Hitachi for its value-line 8mm products.
Optical Disk
Electronic data storage medium that can be written to and read. CD R 650MB 700 MB Write once Read Many Format CD RW Rewritable 650MB 700MB Rewritten as many as 1000 times
DVD
DVD 4.8 GB (Regular DVD) Dual Layer DVD (8.84 GB) Write once Read Many CD format
Blue Ray DVD (BD/DVD) Using a blue violet laser (450nm) which is shorter wavelength than a red laser (600nm) This allows data to be packed more tightly and stored in less space 25GB/50GB
The URL for NetBackup Enterprise Server 6.5 hardware compatibility matrices for specific devices on specific OS platforms is: http://ftp.support.veritas.com/pub/support/products/NetBackup_Enterprise_Server/2 84599.pdf/
SCSI
Small Computer Systems Interface Support for many devices Single SCSI Channel 7 Devices Bus mastering Access to multiple devices TCQ Intelligent BUS MTBF
SCSI
SCSI
SCSI Connector 50pins / 68pins 50 pins 8 bit system
SCSI
SCSI I 5Mb/sec SCSI II 10mb/ Sec (Fast SCSI / Wide SCSI) Single Ended/Differential Versions SCSI III 40Mb/Sec Ultra / Ultra 2 / Ultra 3 / Ultra 320
Termination
SCSI Termination/Cabling
SE Single Ended
Single-ended SCSI uses half of the cable wires for carrying data and control signals and the other half for ground. Because of this configuration, signals on a single-ended bus are more susceptible to corruption due to noise,
SCSI Termination/Cabling
HVD High Voltage Differential (HVD) SCSI uses a set of two wires for each data or control signal on the bus. Each signal is determined by the difference in voltage between the line pair, resulting in higher noise tolerance. This allows for using
SCSI Termination/Cabling
LVD (Low Voltage Differential)
Like HVD, Low Voltage Differential (LVD) SCSI also uses two wires for each signal. This brings the differential bus advantage of reduced signal corruption due to noise. Another advantage of LVD is that it uses lower voltages than HVD, reducing cost and power consumption. Most LVD devices are "multimode", automatically detecting when they are attached to a single-ended bus, and operating in singleended mode. This makes it possible to mix LVD and single-ended devices on the same bus. However, if only one single-ended device is connected to
SCSI - Drawbacks
Additional Hardware Lower Volume Cost
Storage Unit
Definition:- A Storage Unit is a group of one or more storage devices of specific type and density that are attached to a netbackup server. It is the logical destination for client data Tapes,Tapedrives,Robots,Library>Target Library
Configuration of Policies
Policy Definition:- The set of rules that netbackup follows when backing up its clients In Netbackup Administration Console: How to create a new policy - Note: Policy name should not have any spaces What does a Policy define? - Which Client,When/What/Where/How to Backup
Policy
Attributes Schedules Clients Backup Selections Note:-When seen from the activity monitor,the parent backup jobs will have a under schedule but child jobs will have the description,for eg:first backup/daily/weekly/monthly.
Take checkpoints every --minutes ->If a backup job is suspended for some reason and then resuming the job from the most recent time can be possible if the above option is set for example to 15 mins. Limit jobs per policy->2 Job priority ->9999 Eg: DB archive jobs have highest priority.,hence set it to 9999.
All the 3 files are backed up in 3 separate streams on to the tape drive.
bpcd
Catlog
Tape
Files
List of Daemons
Bpcd ->Client Daemon Bpbrm -> recover manager Bptm->Tape manager Bprd->request daemon Bpshed->schedule daemon Bpdbm-> database daemon Bpjob ->Netbackup job daemon Vmd->Volume Manager Daemon Bpbkar(32)
7 Steps
1.When the Daemons wake up interval occurs,the bprd on master will wake up and start the bpsched. Note: bprd is also activated on media server but after checking the bp.conf file(/usr/openv/netbackup/bp.conf) and determining that it is not running on master server, bprd becomes inactive.
2. Bpsched calls bpdbm to check the policy for auto backup schedules
Bpsched calls the child process to handle the backup Bpdbm calls bpjobd to communicate with job catalog 3. Bpsched child contacts the media server to start the bpcd
5.Clients bpcd starts the bpbkar(32) 6.Bpbkar(32) reads data from client Communicates directly with bpbrm to send catalog meta data which is ultimately written to images catalog Send the backup data stream to the child bptm process 7. The child bptm process passes the data to the parent bptm process which writes data directly to the destination media.
Port numbers
Bpcd - > 13782 Bprd -> 13720 Bpdbm -> 13721 bp.conf file is present in all netbackup master/media/client (unix)servers.It contains the host name and ipaddress of all the nodes involved in netbackup
7.Netbackup Error Codes and Trouble shooting Activity Monitor->RC->Fliter:Status>1=>Shows all failed jobs. Common Error Codes: 0->Backup successful 1->Backup Incomplete 11-> System call failed Rerun backup job after ensuring that ping from master to media and client is fine.Else raise a ticket and assign to Intel team
12,13,14-> File Open/Read/Write Failed 21,24->Socket open/write failed ->Restart Netbackup client services 41->Network connectivity timed out 42->Network Read Failed -> If Client cant be pinged then raise a ticket.
usr\openv\veritas\netbacku p\bin\admincmd
Bpmedialist m <Tape id> -h <Media server> {Status of media} Bpmedialist summary {All medias and related info like>expiration date} Bpmedia freeze/unfreeze m <media id> -h <media server> Bpgetconfig g <clientname> -L {Client details} Bperror u <err code> {Lists jobs failed with error code..} Bpdbjobs summary {activity monitor} Bpdbjobs report {Number of jobs queued/running/failed} Bpimagelist U <Client Name> -d <Start date> -d <End Date> {Images of backup for the client}
usr\openv\veritas\volmgr\bin
Vmoprcmd {To check if drives are up or down} Vmoprcmd h <media server> -resubmit/reset <0/1 drive num> Vmquery m <Tape id> {Status of the tape/media} Vmupdate rt <TLD/ACS> -rn <0/1> {To do robot inventory } Vmcheckxxx rt <TLD/ACS> -rn <0/1> {Lists all slots in the tape lib. Can check slot availability} Robtest -> 1<yes> ->s d/s<show all drives/slots> ->m d 1 to s 6 <move media in drive 1 to slot 6> ->m s6 to d 1 <move media from slot 6 to drive 1> Note:-Robtest is done to check if a media is stuck in a drive To check if robot is working fine move tapes from drive to slot or vice versa Vmchange m <media id> -p <pool num> { To move a new media from one vol pool to another }
usr\openv\veritas\netbacku p\bin\goodies
Available_media {Shows all available medias in the environment} Available_media |grep I Full |grep I TLD { To get the details of all the available medias which are full in the tape library} Available_media | grep I <Tape id> {Check if media is present in the tape library.If yes then the result is TLD else the result is none}
Restore
Info needed:-Source server,Dest server,source file path,dest file path,date of backup. Source client,Dest client,Policy type,Start /End date and time,Preview and Restore Tabs Preview shows the medias needed. Check media robot type{TLD/none} for availability of tape. If destination is not given then the root of netbackup will get filled.So be careful.
Health Check
Regular Admin Job to check the activity monitor for errors Coordinate with other teams and get issues resolved. Have the contact details of other teams Raise tickets when required. SEV 1/2/3/4->8hrs/16hrs/5days/10days (varies according to SLA)
Health Check(Contd)
If there is a ticket breach then call for a meeting Put ticket on hold when other teams are handling Check for failed jobs.Usually backup tickets take time to resolve On Call Support ->24x7 support
Imp Questions
How to duplicate tapes? What is a catalog? And wat care should be taken in managing catalogs? What is multiplexing and multistreaming with examples? What is high and low water mark? How to restore backup medias from one master to other master server? If a catalog is corruped? How to restore? How to archive catalog or how to reduce catalog if catalog sixe is increased? How to use netbackup scripting for fine tuning? Basic parameters to create new policy? How is image stored in netbackup? If your last catalog backup was two days ago and the master server crashed what would you have to do to bring your environment back up to present time? There is a Tape library with 10 drives ...Can we able to create 2 Storage units.....?