FORCE_RESTORE_MEDIA_SERVER

[Kennetu]

I forced a restore to use a particular media server with statement:

FORCE_RESTORE_MEDIA_SERVER egypt borno

but still not working fine. It is still using another server instead of borno.

Can anyone help me?

Regards,
Kenneth.

[dltsc]

Have you followed the steps outlined here:

http://seer.entsupport.symantec.com/docs/306838.htm

[dltsc]

Also, does the server borno have access to the media that you are attempting to restore from? And have you put the FORCE_RESTORE_MEDIA_SERVER option in the bp.conf of the master server?  After adding that entry it is necessary to restart the bprd process.

[Kennetu]

Yes, borno can access the media and the FORCE_RESTORE_MEDIA_SERVER option was properly put in the bp.conf of the master server. Also enteries were correctly added, but still not picking borno as the media server.
Any way out of this problem?

Thanks
Kenneth.

[dltsc]

Were the NetBackup services on the master restarted?

[J_H_Is_gone · Accepted solution]

does egypt "own" the tapes with the images you are trying to restore?

If egypt did the backup - then when you go to do the restore borno should do the restore.

But if some other server did the backup and owns the tape  then they will still do the restore.

[Kennetu]

Thanks. The NetBackup services on the master restarted and the problem has been partially resolved ;
That is, three out of the five streams for the restore are now using Borno as the media server,
but the remaining two streams are not restoring anything with the following details:

09/18/2009 11:34:49 - begin Restore
09/18/2009 11:34:51 - number of images required: 1
09/18/2009 11:34:51 - media needed: 003174
09/18/2009 11:34:51 - media needed: 005747
09/18/2009 11:34:51 - media needed: 007356
09/18/2009 11:34:51 - media needed: 003459

I have checked and confirmed from the Veritax Netbackup that the above media were properly loaded, write-protected and inventoried.

How can I get out of this fix as I have limited time to finish the restore and clone the database.

I appreciate your speedy reply.

Kenneth.

[Kennetu]

I am waiting for anybody's contribution oh!
A million thanks.

Kind regards,
Kenneth.

[Marianne]

Hi Kenneth
I have seen this when using FORCE_RESTORE... that bprd goes into a 'hang' state.
If you check processes on the master, you will see a 'bprd -dontfork -mpxmain' .
Quickest solution:
Kill the hung restore.
Stop bprd : bprdreq -terminate
Kill all remaining bprd processes (might have to use kill -9)
Start bprd: initbprd
Restore will now work...

I had a call open with Symantec a couple of years ago for a similar problem on an AIX master server. (Customer used FORCE_RESTORE entries on a daily basis to restore data to standby servers.) After MONTHS of sending logs and endless arguments with Symantec, they managed to trace the problem and issued a new bprd binary.  They never acknowledged this problem in future patch releases, although the fix was included in AIX binaries.

[Rakesh_Khandelw]

How many drives do you have available on this media server? Also,check if any of the media require is currently in use by another restore.

I would also check if the media type defined for these tapes and drive types defined on media server are same.