In brief:
QUERY CONTENT can be used to show what is on the volume.
UPDATE VOLUME can be used to mark the volume as destroyed, then user
restore attempts would get it from the copy pool. RESTORE VOLUME can be
used to restore the primary pool volume from the copy pool.
Please refer to the chapter in the Admin Guide on "Protecting and
Recovering Your Server", section "Restoring Storage Pool Volumes" for
further information.
Regards,
Andy
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
Internet e-mail: storman AT us.eyebm DOT com (change eye to i to reply)
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
Jim Kirkman <jmk AT EMAIL.UNC DOT EDU>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
03/06/2003 13:10
Please respond to jim_kirkman
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: missing volume
Thought I'd see what the list thought about the following (sort of a RFC
as it were).
We've got TSM 4.1.4.2 on OS/390 with a 3494 tape library using VTS
volumes for the backuptapepool (soon to implement native 3590). We use
RMM for tape management and send 3490 tapes offsite for DR.
A user was restoring some files and encountered a failure on one
specific file. Error msg was file not found on server. Server log showed
mount failed. So, turns out that the logical volume could not be mounted
because the physical (stacked) volume was/is not in the robot, and I
checked every frame manually!. A search of syslogs from the time the
file was backed up (last Oct.) turned up that the tape had been ejected
(another mystery to solve, where's the darn cartridge!?) yet the data
was not copied off. IBM has no clue how it could have happened that way,
and obviously neither do we.
So, what do we do with this orphaned data? We can query the volume in
TSM so know what data is on the tape. We could run some selective
backups on required data (there's a reasonable number of folders
affected) that would expire these older versions, but that's a bit
arduous. We could delete the volume and discarddata and have it back up
again, but the fact that we discovered this because of the need to
restore makes me a little queasy on that one. We could mark the volume
as damaged, but not sure exactly how that would work. Would the offsite
copies be called back and new copies made, or not? I don't think the
audit volume process would do anything for us.
The other issue is how do we find out what else might have been on this
tape?
What say you?
--
Jim Kirkman
AIS - Systems
UNC-Chapel Hill
966-5884
|
