cancel
Showing results for 
Search instead for 
Did you mean: 

Enterprise Vault backup

mahmoud_rady
Level 3
Partner Accredited

 

Hi

    I need to backup Enterprise vault ( mail server & file server ) in a cluster way components of EV distributed on different cluster machines

 

-  we have 2 SQL servers clustered

- 2 file servers clustered

- 2 mail server clustered  (exchange)

 

netbackup server 7.5 and EV 10

 

Do any one have a simple doc to implement this project

 

Thanks in advance

 

 

10 REPLIES 10

Will_Restore
Level 6

http://www.symantec.com/business/support/index?page=content&id=TECH176623

 

Recommended basic backup configurations when protecting Enterprise Vault 10 using the NetBackup 7.1 and later Enterprise Vault Agent

 

The NetBackup 7.1 and later Enterprise Vault Agent is capable of automatically discovering and protecting the following Enterprise Vault 10 entities:

 

Enterprise Vault 10

  

  • Open Partitions
  • Closed Partitions
  • Ready Partitions
  • Index Location
  • Vault Store databases
  • Directory database
  • Monitoring database
  • Fingerprint database
  • FSAReporting Database
  • Auditing Database


The NetBackup 7.1 and later Enterprise Vault Agent includes 4 different sets of Enterprise Vault Directives; one specific to Enterprise Vault 7.5, one specific to Enterprise Vault 8.0, one specific to Enterprise Vault 9.0, and one specific to Enterprise Vault 10.

 

 

Because of changes to the logical architecture introduced in Enterprise Vault 8.0 and the introduction of the logical container "Vault Store Group", the way the Netbackup 7.1 and later Enterprise Vault Agent enumerates and backs up Enterprise Vault is granular and per individual entity. The following Directives are available specific to Enterprise Vault 10:


EV_OPEN_PARTITION=<VAULT_STORE_NAME>

EV_CLOSED_PARTITIONS=<VAULT_STORE_NAME>

EV_READY_PARTITIONS=<VAULT_STORE_NAME>

EV_INDEX_LOCATION=<EV_SITE_NAME>

EV_VAULT_STORE_DB=<VAULT_STORE_NAME>

EV_FINGERPRINT_DB=<VAULT_STORE_GROUP_NAME>

EV_DIR_DB (No target necessary)

EV_MONITORING_DB (No target necessary)

EV_AUDIT_DB (No target necessary)

EV_FSAREPORTING_DB (No target necessary)

 

For a detailed explanation on the above EV 10 Directives please refer to the "Symantec NetBackup (tm) 7.1 for Enterprise Vault Agent Administrator's Guide for Windows":    http://www.symantec.com/docs/DOC3660

 

NOTES!

  • The FSAReporting Database only gets created when File System Archiving is configured and then only if FSAReporting is enabled and configured.
  • The Auditing Database only gets created if Enterprise Vault Auditing is enabled and configured.


In order to back up the Enterprise Vault 10 entities in a sequence that allows restores to be performed that require a minimum of manual re-synchronization, configure NetBackup Policies according to the following recommendations.

In the Enterprise Vault Policies listed below, it is recommended that the Policy Client is the Alias of an Enterprise Vault Server belonging to the Enterprise Vault Site.

 

a. Create an Enterprise Vault Policy that has the EV_OPEN_PARTITION and EV_FINGERPRINT_DB as it's Directives. Within this Policy, create 2 Schedules, 1 Full backup and 1 Differential Incremental Backup.

b. Create an Enterprise Vault Policy that has the EV_DIR_DB, the EV_MONITORING_DB, (if configured) the EV_AUDIT_DB and (if configured) the EV_FSAREPORTING_DB as it's Directives. Within this Policy, create 2 Schedules, 1 Full Backup and 1 Differential Incremental Backup.

c. Create an Enterprise Vault Policy that has the EV_INDEX_LOCATION as it's Directive. Within this Policy, create 2 Schedules, 1 Full backup and 1 Differential Incremental Backup.

d. Create an Enterprise Vault Policy that has the EV_CLOSED_PARTITIONS and EV_READY_PARTITIONS as it's Directives. Within this Policy, create 2 Schedules, 1 Full backup and 1 Differential Incremental Backup.

 

For each of the above Policy Schedules configure their respective Start Windows so they execute in the order set out below. Adjust the Start Times according to individual EV Site configurations and size of entities. Configure the Start Windows to take into account the duration of the backups so that none of the backups overlap each other.

 

Configure Full and Differential Incremental schedules to take into account the size of the EV environment and execute the schedules for Full and Differential Incremental in the sequence they are listed here i.e. run Full backups once per week and Differential Incremental backups daily the rest of the week or the below sequence daily if possible.

NOTE! It is very important that the Database and Open Partition backups include Differential Incremental Schedules since this schedule type is the only one that will back up the Database Transaction Log and truncate it. If the Transaction Log is not truncated it will continue to grow in size and could eventually lead to a disk full situation.

 

For a very large EV environment, consider running Full backups monthly, Cumulative Incremental backups weekly (add Cumulative Incremental Schedules to the Policies created above) and Differential Incremental backups daily. In addition, backups of non-critical entities such as Closed and Ready Partitions can be scheduled to run less frequently.  

 

1st: The Full Schedule of the EV_OPEN_PARTITION and EV_FINGERPRINT_DB in the EV Policy created in a. above.

2nd: The Full Schedule of the EV_DIR_DB, the EV_MONITORING_DB, (if configured) the EV_AUDIT_DB and (if configured) the EV_FSAREPORTING_DB in the EV Policy created in b. above.

3rd: The Full Schedule of the EV_INDEX_LOCATION in the EV Policy created in c. above.

4th: The Full Schedule of the EV_CLOSED_PARTITIONS and EV_READY_PARTITIONS in the EV Policy created in d. above

(If running Full backups weekly or monthly, run the following Cumulative Incremental backups weekly and/or Differential Incremental backups daily in the sequence set out here)

5th: The Differential Incremental Schedule of the EV_OPEN_PARTITION and EV_FINGERPRINT_DB in the EV Policy created in a. above.

6th: The Differential Incremental Schedule of the EV_DIR_DB, the EV_MONITORING_DB, (if configured) the EV_AUDIT_DB and (if configured) the EV_FSAREPORTING_DB in the EV Policy created in b. above.

7th: The Differential Incremental Schedule of the EV_INDEX_LOCATION in the EV Policy created in c. above.

8th: The Differential Incremental Schedule of the EV_CLOSED_PARTITIONS and EV_READY_PARTITIONS in the EV Policy created in d. above

 

mahmoud_rady
Level 3
Partner Accredited

Hi  wr

   thanks for the fast response but my question is how to configure netbackup for EV if its DBs on 2 SQL server & 2 file servers & 2 mail servers (distributed components)

How can i do such this implementation

 

Thank you

 

Yasuhisa_Ishika
Level 6
Partner Accredited Certified

Install NetBackup clients on EV servers and SQL Server nodes, then configure backup policy as wr suggested. 

If you get status code 39, please try to change client name on SQL nodes to cluster virtual name. Even if this technote and Administration Guide does not say it is required on NetBackup 7.5, I doubt this is also required in NetBackup 7.5.

http://www.symantec.com/docs/HOWTO54645

Mark_Solutions
Level 6
Partner Accredited Certified

How much data do you have in EV?

As you are doing mail and file system it sounds like you may have quite a lot!

If that is the case then you may find the EV Agent a little too slow for you.

Stepping away from the EV policy for a moment .. lets look at you components ...

You have a clustered file server to back up - the important thing here is to not be running your archiving window at the same time as your backup - so let it do its archiving run first and then once that window closes run your backup - at this point EV's work is done so you get a consistent state (remember to add bpbkar32 as a process for EV to ignore on the file server so that the backup does not recall all of your files) - the timing then is not important

Next you have your exchange servers - same applies here so keep the archiving run earlier than the backup so that all the EV work is done before your backup runs - the timing then is not important

Next you have a SQL Cluster - these hold the EV Databases so you want to back these up at the same time as EV itself (the indexes and vault partitions) so schedule them to run after the EV backup starts (should be failrly easy as the EV backup will always take a lot longer to run than the SQL backups so you should have plenty of time to play with here.

Finally you have the EV Server backup - start this after the EV mail and file system archive runs have finished but before your SQL Server backup starts. You can use the EV agent if you wish, but just for Index and Vault Store partitions or you may choose to stream the backups as normal windows backups or better still as Flashbackup-Windows policies to get then to run at a good rate.

As a standard of flashbackup type you need to put EV into backup mode during these backups so may need to use bpstart_notify / bpend_notify scripts to do that or if streaming you may need parent start notify scripts running from the Master (not so easy)

If using flashbackup you will also need to add trigger file creation after the backup completes so thet EV knows it has been backed up.

Sounds a lot and a little complex but as long as you get the archiving run done first then when you backup the file server and Exchange doesn't really matter - the important bit is backing up EV itself in backup mode and getting the SQL backups doen at the same time

Hope this all makes sense and helps

 

mahmoud_rady
Level 3
Partner Accredited

Hi All

    We created the policy and configured it now the DBs backup policy completed successfuly but the index is failing on the file server and on mail server cluster is starting with 2 jobs the first is waiting for retry for  time and the second job is completing with the EV index backup for file server any ideas i can check

Mark_Solutions
Level 6
Partner Accredited Certified

What exactly have you configured?

We will need to see your policy(ies) to have an idea of what is happening

mahmoud_rady
Level 3
Partner Accredited

Hi all

 

  I found that our EV is version 9 and netbackup agent 7.5.0.1

  is there any configuration specific for EV 9

 

I heared that for a successful backup for EV version 10 i must exclude some files from exclude list is this for version 9 also

 

thanks

Yasuhisa_Ishika
Level 6
Partner Accredited Certified

It sounds like you configured backup without Enterise Vault agent. Right?

Please show us your poclicy configurations you created to protect EV.

mahmoud_rady
Level 3
Partner Accredited

Hi all

 

   the site is  

-  we have cluster SQL for DBs

- and cluster EV 9 for file server and its index on EMC centera configured as a shared partition for file server and its index is 8 folders

- and cluster EV 9 for mail server and its index on EMC centera configured as a shared partition for mail server index is 8 folders

 

my backup is consisting of 4 policies

- first one for open partition & fingerprint DB and is completing successfully with status 1 as some files be in use

- second one is for dir db & monitoring Db and is completing successfully

- third one is for index and here is the problem the backup starts with parent job and then it opens child job and another child job and then snapshot job and then 8 jobs for 8 folders index the 8 jobs for 8 folders index all complete successfuly and snapshot complete successfully and then here is the problem the parent job for snap is failling with status 2 ( none of the requested files were backed up ) after retrying and when going to list the backups from from BAR we found that indexs are backed up

 

the question is is there any special configuration specific for EV 9 like exclude lists or farther configuration that i missed specific for EV

 

thank you

Yasuhisa_Ishika
Level 6
Partner Accredited Certified

IT seems like NetBackup for EV agent backups... No exclude list settings for such backup type.

I'm not so familiar with EV agent for now, so it is hard for me to help you explicitly. If you never seen, please check Best Practice doc below.
http://www.symantec.com/docs/TECH147148

BTW, it is better to show us output of "bppllist policy_name -U" command and detail status of failed job for further help.