[Mauiusers] Does maui have a hard-coded walltime limit???
Thomas Dargel
td at chemie.hu-berlin.de
Tue Mar 20 05:06:54 MDT 2007
Dear mauiusers,
please, can somebody give me a hint what's going wrong here:
One job was cancelled by maui with the following log-message:
03/10 00:08:00 MRMWorkloadQuery()
03/10 00:08:00 MPBSWorkloadQuery(node01,JCount,SC)
03/10 00:08:00 MPBSJobUpdate(30766,30766.cnode01.mauicluster,TaskList,0)
03/10 00:08:00 MStatUpdateActiveJobUsage(30766)
03/10 00:08:00 MResDestroy(30766)
03/10 00:08:00 MResChargeAllocation(30766,2)
03/10 00:08:00 MResJCreate(30766,MNodeList, -INFINITY,ActiveJob,Res)
.
.
.
03/10 00:08:00 INFO: 20 PBS jobs detected on RM node01
03/10 00:08:00 INFO: jobs detected: 20
03/10 00:08:00 MStatClearUsage(node,Active)
03/10 00:08:00 MClusterUpdateNodeState()
03/10 00:08:00 INFO: requeue value 208046109.00 found for immediate action (T: 00:00:00)
03/10 00:08:00 INFO: requeue value 208076658.00 found at completion of job 30766 (T: -00:09:59)
03/10 00:08:00 MQueueSelectAllJobs(Q,HARD,ALL,JIList,DP,Msg)
03/10 00:08:00 INFO: job '30766' Priority: 252783
03/10 00:08:00 INFO: Cred: 0(00.0) FS: 244279(00.0) Attr: 0(00.0) Serv: 8504(00.0) Targ: 0(00.0) Res: 0(00.0) Us: 0(00.0)
.
.
.
.
03/10 00:08:29 MRMWorkloadQuery()
03/10 00:08:29 MPBSWorkloadQuery(node01,JCount,SC)
03/10 00:08:29 MPBSJobUpdate(30766,30766.cnode01.mauicluster,TaskList,0)
03/10 00:08:29 MStatUpdateActiveJobUsage(30766)
03/10 00:08:29 MResDestroy(30766)
03/10 00:08:29 MResChargeAllocation(30766,2)
03/10 00:08:29 MResJCreate(30766,MNodeList, -INFINITY,ActiveJob,Res)
.
.
.
.
03/10 00:08:29 MStatClearUsage(node,Active)
03/10 00:08:29 MClusterUpdateNodeState()
03/10 00:08:29 INFO: requeue value 208044630.00 found for immediate action (T: 00:00:00)
03/10 00:08:29 INFO: requeue value 208076658.00 found at completion of job 30766 (T: -00:10:28)
03/10 00:08:29 MQueueSelectAllJobs(Q,HARD,ALL,JIList,DP,Msg)
03/10 00:08:29 INFO: job '30766' Priority: 252783
03/10 00:08:29 INFO: Cred: 0(00.0) FS: 244279(00.0) Attr: 0(00.0) Serv: 8504(00.0) Targ: 0(00.0) Res: 0(00.0) Us: 0(00.0)
.
.
.
.
03/10 00:08:29 ALERT: job '30766' in state 'Running' has exceeded its wallclock limit (8639999+S:0) by 00:10:28 (job will be cancelled)
03/10 00:08:29 MSysRegEvent(JOBWCVIOLATION: job '30766' in state 'Running' has exceeded its wallclock limit (8639999) by 00:10:28 (job will be cancelled) job start time: Wed Nov 29 23:58:02,0,0,1)
03/10 00:08:29 MSysLaunchAction(ASList,1)
03/10 00:08:29 MRMJobCancel(30766,MOAB_INFO: job exceeded wallclock limit ,SC)
03/10 00:08:29 MPBSJobCancel(30766,node01,CMsg,Msg,MOAB_INFO: job exceeded wallclock limit)
03/10 00:08:29 INFO: job '30766' successfully cancelled
There is no walltime limit set, neither at torque/maui nor in the job script.
Where does the 'wallclock limit (8639999)' come from??
Is there a hardcoded limit in maui??
Any help is appreciated,
thank you in advance
Thomas Dargel.
More information about the mauiusers
mailing list