[torqueusers] 1.1.0p6 cpu time counter fails with very long jobs?

Mikko Huhtala mhuhtala at abo.fi
Thu Feb 3 01:50:00 MST 2005


Mikko Huhtala writes:
 > Garrick Staples writes:
 >  > Do you have anything in your mom logs at that time?

The pbs_mom logs are appended below. The jobs seem to simply have
exceeded their cput limits (cput used is bogus).

Mikko

----------------------------------------------------------------------


root volvox7:/var/spool/pbs/mom_logs > grep 21[45].volvox.abo.fi *
20050105:01/05/2005 19:44:54;0001;   pbs_mom;Job;PMomFinalizeJob3;job 214.volvox.abo.fi started, pid = 30328
20050105:01/05/2005 19:44:54;0008;   pbs_mom;Job;214.volvox.abo.fi;Job Modified at request of PBS_Server at volvox.abo.fi
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;cput 19187329 exceeded limit 15552000
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;kill_task: killing pid 30328 task 1 with sig 15
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;kill_task: killing pid 30343 task 1 with sig 15
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;kill_task: killing pid 30344 task 1 with sig 15
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;kill_task: killing pid 30350 task 1 with sig 15
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;scan_non_child_tasks;found exited session 30328 for task 1 in job 214.volvox.abo.fi
20050130:01/30/2005 16:10:31;0008;   pbs_mom;Job;214.volvox.abo.fi;Terminated

root volvox8:/var/spool/pbs/mom_logs > grep 21[45].volvox.abo.fi *
20050105:01/05/2005 19:56:58;0001;   pbs_mom;Job;PMomFinalizeJob3;job 215.volvox.abo.fi started, pid = 29254
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;cput 19187337 exceeded limit 15552000
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;kill_task: killing pid 29254 task 1 with sig 15
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;kill_task: killing pid 29269 task 1 with sig 15
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;kill_task: killing pid 29270 task 1 with sig 15
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;kill_task: killing pid 29276 task 1 with sig 15
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;scan_non_child_tasks;found exited session 29254 for task 1 in job 215.volvox.abo.fi
20050130:01/30/2005 19:43:13;0008;   pbs_mom;Job;215.volvox.abo.fi;Terminated



More information about the torqueusers mailing list