[torqueusers] qsub working only for _some_ users??
Gerolf Ziegenhain
mail.gerolf at ziegenhain.com
Thu Sep 27 06:57:30 MDT 2007
Hi everybody,
I'm administrating a small cluster and installed torque & maui recently. The
jobs can be submitted from a head node and on all slave nodes nis&nfs&ssh is
working nicely. Just now I discovered a strange problem: With some users I
can submit jobs but with others submitting jobs failes with:
************* Failure *************
qsub -I
qsub: waiting for job 4772.wap.physik.uni-kl.de to start
qsub: job 4772.wap.physik.uni-kl.de apparently deleted
************* Failure *************
There are still enough nodes free...
************* maui.cfg *************
SERVERHOST wap
ADMIN1 root
RMCFG[base] TYPE=PBS
RMPOLLINTERVAL 00:00:30
SERVERPORT 42559
SERVERMODE NORMAL
LOGFILE maui.log
LOGFILEMAXSIZE 10000000
LOGLEVEL 3
QUEUETIMEWEIGHT 1
BACKFILLPOLICY FIRSTFIT
RESERVATIONPOLICY CURRENTHIGHEST
NODEALLOCATIONPOLICY MINRESOURCE
GROUPCFG[DEFAULT] MAXPROC=56 MAXMEM=4096 MAXNODES=14
USERCFG[DEFAULT] MAXPROC=50
************* maui.cfg *************
There are different users and groups configured in NIS. All of them can log
in nicely everywhere. But for some _specific_ users submitting doesn't work.
I've already tried to restart the head/nodes (NIS update etc) and looked at
/var/log/... without finding anything special.
What could be a point to look at?
Best regards:
Gerolf
--
Dipl. Phys. Gerolf Ziegenhain
Office: Room 46-332 - Erwin-Schrödinger-Str.46 - TU Kaiserslautern - Germany
Web: gerolf.ziegenhain.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20070927/f77f20b5/attachment.html
More information about the torqueusers
mailing list