SQL Transaction Log terminating with error 2 'none of the requested files were backed up'

[Nerdling]

Hi Guys,

I'm running NBU on a Clustered SQL server.  When running on Host A backups completle sucessfully however, when running on Host B they fail with the following error: none of the requested files were backed up(2)

Below is the job log...

30/06/2011 11:56:39 PM - requesting resource HO-MSL6000-HCART3-Group
30/06/2011 11:56:39 PM - requesting resource honbums.domain.hosted.NBU_CLIENT.MAXJOBS.HostB
30/06/2011 11:56:39 PM - requesting resource honbums.domain.hosted.NBU_POLICY.MAXJOBS.HO-Clus10--Clus20-Transaction-logs
30/06/2011 11:56:39 PM - granted resource honbums.domain.hosted.NBU_CLIENT.MAXJOBS.HostB
30/06/2011 11:56:39 PM - granted resource honbums.domain.hosted.NBU_POLICY.MAXJOBS.HO-Clus10--Clus20-Transaction-logs
30/06/2011 11:56:39 PM - granted resource honbums-hcart3-robot-tld-0
30/06/2011 11:56:39 PM - estimated 0 Kbytes needed
30/06/2011 11:56:40 PM - started process bpbrm (7828)
30/06/2011 11:56:40 PM - connecting
30/06/2011 11:56:43 PM - connected; connect time: 00:00:03
1/07/2011 1:23:18 AM - end writing
none of the requested files were backed up(2)

[Marianne]

Are you using the virtual hostname in the backup policy?

Is this virtual hostname the SQLHOST in the script?

Please post backup script as well as dbclient log on failing node.

[Nerdling]

>>Are you using the virtual hostname in the backup policy?

Yes

>>Is this virtual hostname the SQLHOST in the script?

Don't know.  How would I check?

>>Please post backup script as well as dbclient log on failing node.

Where would I get these from?

[Marianne]

The script location can be found in the policy Backup Selection tab. If the script is on local disk (i.e. C:\Program Files\....) it needs to be copied to all cluster nodes. I normally recommend that customers save the script in a shared folder.

The log directories can be found under <install_path>\veritas\netbackup\logs. They don't exist by default. If no directories, create bpcd (for connection troubleshooting) and dbclient (for database backup).

[Oddy]

If it works when the database is located on node A but fails when it is on node B

Try to login to node B as a local administrator with sql sysadmin rights, open the Netbackup MSSQL Database gui and put the sql cluster name as the sql host and try to kick off the backup from the client side.

PS ! If your sql server is windows 2008 and you cannot get it to connect to the sql host, try to Right Click on the Netbackup MSSQL database icon and select "Run as administrator".

I.e go to File, Backup..., select a small database in the window on the right hand side and save the script.

Say Yes to open the script in notepad and make sure that the BROWSECLIENT is the sql cluster name, excactly the same name as what you have in put in the sql policy on the master. If the line is not there, add it under the NBSERVER line.

Create log directories on the sql node B server.

<install dir>\netbackup\logs\bpcd

<install dir>\netbackup\logs\dbclient

Go to File, Manage script files and kick off the batch script you created. Check in the Activity monitor on the master if the backup worked and in the View Status window on the client.

If the backup is successful, try to change the netbackup client service on node B to run in the same local admin account you are logged in as and try to kick off the backup from the master server.

If it fails, check the dbclient log for errors.

I hope this helps.

/Oddy