I'm running into something similar with incremental backups. The backups
start normally, proceed for an hour or so, and then hang. I get these
errors in the error log

12/11/2009 09:43:32 cuGetBackQryResp: Server aborted txn, reason: 1

Even restarting the dsmc processes doesn't get anything to happen, but
db2sysc on the server is busy churning away on the CPU. I've got a PMR
open now, and I'll let you know what I find out.

Zoltan Forray/AC/VCU wrote:

..........when using a V6.1 server.

I just stumbled upon a new issue with a V6.1 server.

As you recall, in my current saga of the trials-and-tribulations of
attempting to restore a large backup/volume (F: drive) to a Windows client
(20M+ files),  I found that I can not do a NQR since I must process
inactive files since most everything on this volume has gone MIA.  One
other volume (G: drive) is fine but still needs to be restored since both
volumes are in an array that needs to be wiped/restored.

Over a month ago, I had attempted to move this nodes backups to my V6.1
TSM server, only to have the V6.1 system crash numerous times just during
the export/import process.  Also, the server crashed numerous times just
trying to perform incremental backups for this client machine.  So, in
disgust, I returned this node to the 5.5 server and the backups ran
normally.  Since this time, the 6.1 server has been upgrade to 6.1.2.1 but
I never went back.

Yesterday, I decided to try to perform an NQR from the old backups for
this node, using the 6.1.2.1 server, since those backups, while old, are
all active.  This way, I would also be able to figure out where the
majority of the files are so when I went back to the 5.5 server to restore
to a more current date, I would have some idea of what directories to
process in small chunks.

Well, this has not been going very well and brings out another issue with
V6.1 servers.

When I restored the G: drive using the 5.5 server and NQR, the restores
started to flow after about 3-5 minutes, allowing for tape mounts and
such.  Went rather well,  especially with 5-restore streams and 500GB+/27M
objects.

With the 6.1 server and the F: drive NQR (also upgrade the client to the
latest 6.1.2.0),  from the time I clicked on restore to when the restores
actually started, has been over 4-hours.  No tape activity.....no session
activity.........nothing...........

So, what has the TSM server been doing for the past 4-hours?

--
-- Skylar Thompson (skylar2 AT u.washington DOT edu)
-- Genome Sciences Department, System Administrator
-- Foege Building S048, (206)-685-7354
-- University of Washington School of Medicine