Online backup fails with sql0911

[andrejf]

Occasionally, regular online backup fails with sql0911 reason 68,
"lock-timeout" seconds after the command is issued.
Normally a second try is successfull. The database is always in use.

Can someone technically explain why this happens, which locks (type,
scope) prevent a backup to start, and maybe, is there a workaround
(not affecting the applications).

db is on a 8.2 fp11 level, userexit on, i am backuping online to tsm.

Regards,
AF

[shen...@gmail.com]

You must have somebody exclusive locked some table/row, backup wait
until itself timeout.
try set the LOCKTIMEOUT higher, so that backup will wait longer for
your other application to finish its work.

[dotyet]

but that value is set at DB CFG level. It will be applicable across
application, or am I missing something?

rgds,
dotyet

[dean.c...@gmail.com]

On Oct 24, 7:58 pm, andrejf <andrej.furla...@gmail.com> wrote:
> Occasionally, regular online backup fails with sql0911 reason 68,
> "lock-timeout" seconds after the command is issued.
> Normally a second try is successfull. The database is always in use.

Does the table have xLOBs in it?

[andrejf]

On 27 okt., 04:46, "dean.cochr...@gmail.com" <dean.cochr...@gmail.com>
wrote:

Not in user tables (there are some in sysibm.*). But we use some long
varchars...

[deangc]

> > Does the table have xLOBs in it?
>
> Not in user tables (there are some in sysibm.*). But we use some long
> varchars...

I don't think it's the same issue, then, sadly. We had a number of
clients with this issue. There are cases where locks will be held in
tables with LOBs in them in SMS tablespaces, and these can cause
online backups to fail. I understand that this has been fixed in 9.x,
but is still an issue in 8.x. This is not an issue if you are using
DMS tablespaces.