[torqueusers] Some questions on elipogue/prologue

Garrick Staples garrick at usc.edu
Tue Jun 5 10:37:58 MDT 2007


On Tue, Jun 05, 2007 at 01:48:44PM +0300, Alexander Piavlo alleged:
> Hi Garrick,
> 
> >> And what happens then
> >> (a) a job fails to stagein files before start
> >> and is put W state to later be resubmited if it is rerunable.
> >> (b) if job fails during execution
> >
> >In both cases epilogue scripts are run.
> 
> Well, for me, in case (a) none of the epilogue scripts run.
> My current problem is:
> at /var/spool/pbs/mom_priv/config i've added:
> $tmpdir /scratch
> Then a job is submited a /scratch/${PBS_JOBID} is created. And i have
> a epilogue &
> epilogue.precancel scripts to remove this dir, this works ok on job
> completion or then job canceled.
> But if a job has a stagein specified with -W and the staging fails the
> jobs is but in W state
> to be later resubmited, the /scratch/${PBS_JOBID} is created, but none
> of the epilogue
> scripts are run to delete this dir.

Why do you remove it in epilogue?  If pbs_mom created it, it will remove
it.

-- 
Garrick Staples, GNU/Linux HPCC SysAdmin
University of Southern California

Please avoid sending me Word or PowerPoint attachments.
See http://www.gnu.org/philosophy/no-word-attachments.html
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/torqueusers/attachments/20070605/220175c2/attachment.bin


More information about the torqueusers mailing list