If you suspect any corruption to your TSM database, I can't think of a
better first step than to call TSM Support and work with someone there to
isolate the issue.  There are switches to the AUDIT DB command that can
greatly reduce the running time, but you have to work through with Support
to identify if they're relevant for you, and which ones to use.

Nick Cassimatis

----- Forwarded by Nicholas Cassimatis/Raleigh/IBM on 11/03/2006 01:48 PM
-----

"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 11/03/2006
11:12:36 AM:

> We are trying to restore a filesystem on an Solaris server and are
> experiencing the following error messages, over and over:
>
> 11/2/2006 6:10:50 PM ANR9999D afcputil.c(1456): ThreadId Volume 17990 is
> not defined for pool -2.
>
> I did some digging and the only related hit I found on the ITSM site
> refers to APARS that are 2-3 years old
>
> IC42603: DELETE VOLUME DISCARDDATA=YES MAY LEAVE ORPHANED ENTRIES IN THE
> ITSM SERVER DB
>
> Yes, I have been doing some cleanups of old copypool volumes (recreated
> all offsite copypools to LTO) and am doing DELETE VOLUME DISCARDDATA=YES
> on the unneeded volumes.
>
> I have a concern that this (and other similar APARS) are dated 2004/2005.
> My TSM server is at 5.3.3.2. Another problem I have is the APARS all
> discuss ARCHIVE data. We are trying to restore from backups, not
archives.
>
> What concerns me more about these symptoms/errors is a disscussion I read
> on ADSM.ORG, posted within the last 2-weeks
> (http://my.adsm.org/modules.php?
> op=modload&name=phpBB_14&file=index&action=viewtopic&topic=8179)
> , with the comment:
>
> "To fix the problem I will have to audit my database. I was told they
> found a problem with the audit DB utility in the current version of TSM,
> so I'm waiting for the next release before we do the audit. "
>
> but no other details on what TSM level.
>
> I don't look forward to a 2-3+ days outage to perform an AUDITDB (here
> again, I am lacking details on how much of an audit I need to perform).
>
> I was about to upgrade to 5.3.4 but don't think I should disturb the
> environment until this issue is resolved, unless the 5.3.4 server level
> fixes the problem.
>
> Looking for suggestions/ideas/thoughts on this problem !