[GE users] debugging scheduling issues

templedf dan.templeton at sun.com
Mon Mar 30 15:03:20 BST 2009


I presume everything was working before?  What is your scheduler config 
(qconf -ssconf)?  What does qstat -j say about the job while it's waiting?

Daniel

leonardz wrote:
> On sge 6.0u8 I shut down and restarted and am seeing some very wierd scheduling issue.
>
> with > 200 nodes, almost completely empty I submit a test job, and there are no queued jobs waiting. The job runs in < 1 second and finishes correctly.
>
> Yet it takes > 6 minutes (27 scheduling periods) to run:
>
> qsub_time    Mon Mar 30 09:38:02 2009
> start_time   Mon Mar 30 09:44:49 2009
> end_time     Mon Mar 30 09:44:49 2009
> granted_pe   NONE
> slots        1
> failed       0
> exit_status  0
> ru_wallclock 0
> ru_utime     0
>
>
> How do I find out why this job waits so long to be run?
>
> ------------------------------------------------------
> http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=149292
>
> To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].
>

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=149299

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list