[prev in list] [next in list] [prev in thread] [next in thread] 

List:       grid-engine-users
Subject:    Job wait for ever on a solaris86 cluster running linux OS
From:       "Korambath, Prakashan" <ppk () ats ! ucla ! edu>
Date:       2003-01-23 18:00:28
Message-ID: E120A6DE11CD4149AD9F2F9AE610FDA506B9AA () COLLIE ! ats ! ucla ! edu
[Download RAW message or body]


I submitted a job on a solaris86 cluster running linux OS.  The job
seems to be waiting for ever.

Qstat -j <jobid>

	shows the queue "bass.q" dropped because it is full

Also it didn't seems to have tried on any of the other queues.  There is
practically nothing running on the
Machine where it says it is full. There is no extra output in the
"messages" file in the qmaster directory.

	If we reboot the machine first job get submitted and rest of the
jobs wait for ever.  Looks likes something is missing.  Since we
couldn't get qmon running on that machine, we need to debug by text
commands.

	I was wondering how to debug such a situation.  I couldn't find
anything wrong in qconf -se commands.
Thanks for any help.


Prakashan

[prev in list] [next in list] [prev in thread] [next in thread]