Yeah... it seems to reach walltime limit.. what he mentioned
additionally is there are no .o and .e files left over.
Perhaps, his code doesn't output anything ?
Thanks,
--------------
Robin
[log in to unmask]
513-529-1483
"Academia politics is the most vicious precisely because the stake is
so small" - Kissinger
On Apr 5, 2006, at 3:09 PM, David Woods wrote:
> From the logs, it looks like both jobs ran on compute-1-7.
>
> Looking at the logs on compute-1-7 in /var/spool/PBS, I don't see
> anything
> unusual - but I'm not sure what something "bad" would look like.
>
> It looks like the jobs ended around noon today - 2208 at 12:48 and
> 2209 at
> 11:49. The logs show a Type 54 (PBS_BATCH_Copyfiles) and type 6
> (PBS_BATCH_DeleteJob) request from the PBS server at the same time
> as the
> jobs ended. The entries in the logs like just like other job
> ending entries
> I found in the logs. The PBS temp directories for the jobs are
> gone from
> /tmp so it looks like PBS cleaned up after the jobs.
>
> Dave
>
> -----Original Message-----
> From: Research Computing Support
> [mailto:[log in to unmask]] On
> Behalf Of Robin
> Sent: Wednesday, April 05, 2006 2:51 PM
> To: [log in to unmask]
> Subject: Fwd: Vanishing jobs ?
>
> All,
>
> Logs are at /var/spool/PBS/server_logs
>
> 20060404:04/04/2006 07:14:16;0100;Server@mulnx31;Job;
> 2208.mulnx31;enqueuing into serial, state 1 hop 1
> 20060404:04/04/2006 07:14:16;0008;Server@mulnx31;Job;2208.mulnx31;Job
> Queued at request of [log in to unmask], owner =
> [log in to unmask], job name = Minim81s6, queue = serial
> 20060404:04/04/2006 07:14:16;0008;Server@mulnx31;Job;2208.mulnx31;Job
> Run at request of [log in to unmask] on hosts
> compute-1-7:ncpus=1
> 20060404:04/04/2006 07:14:16;0008;Server@mulnx31;Job;2208.mulnx31;Job
> Modified at request of [log in to unmask]
> 20060405:04/05/2006 12:48:44;0080;Server@mulnx31;Job;
> 2208.mulnx31;Obit received
> 20060405:04/05/2006 12:48:44;0010;Server@mulnx31;Job;
> 2208.mulnx31;Exit_status=0 resources_used.cpupercent=98
> resources_used.cput=29:27:32 resources_used.mem=10588kb
> resources_used.ncpus=1 resources_used.vmem=98496kb
> resources_used.walltime=29:34:28
> 20060405:04/05/2006 12:48:44;0100;Server@mulnx31;Job;
> 2208.mulnx31;dequeuing from serial, state 5
>
> Hmm.. thoughts ?
>
> Thanks,
> --------------
> Robin
> [log in to unmask]
> 513-529-1483
>
> "Academia politics is the most vicious precisely because the stake is
> so small" - Kissinger
>
>
>
> Begin forwarded message:
>
>> From: James Moller <[log in to unmask]>
>> Date: April 5, 2006 2:03:31 PM EDT
>> To: [log in to unmask]
>> Subject: Vanishing jobs ?
>> Reply-To: Research Computing Support <[log in to unmask]>,
>> James Moller <[log in to unmask]>
>>
>> Some of my jobs (viz, Minim81s6 - possible job id 2208 and
>> Minim81s7 - possible job id 2209) appear to have vanished while
>> running this AM. The wall time does not appear to have been
>> reached. Also, no "dot o" file was left to advise of the outcome of
>> the job.
|
