[torqueusers] qdel jobs are not cleanup

"Mgr. Šimon Tóth" SimonT at mail.muni.cz
Wed Sep 15 08:23:09 MDT 2010


> Thanks for suggestions.
> I am not using -p options and they are not mpi jobs.
> 
> I only come to know the job stuck on node when /var space is alarming.
> I would be nice if mom having any parameter who could kill such stuck job.

As I already said, you can cleanup jobs from nodes using the momctl tool.

If the node doesn't report the job at all:
status = .....jobs=624.torque1.ics.muni.cz.....

Then it means only one thing, you spawned a process, but didn't register
it with the node, therefore when the node kills your job, it doesn't
kill the process (because she has no idea that it belongs to the job).

-- 
Mgr. Šimon Tóth

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3366 bytes
Desc: S/MIME Cryptographic Signature
Url : http://www.supercluster.org/pipermail/torqueusers/attachments/20100915/c601612e/attachment.bin 


More information about the torqueusers mailing list