[torqueusers] Negative walltime reported?

Steve Traylen steve.traylen at cern.ch
Fri Feb 6 02:54:22 MST 2009


  We are seeing some jobs with accounting records.

11/17/2008 14:48:38;E;377258.shed.grid.kiae.ru;user=alicesgm group=alice
jobname=STDIN queue=alice ctime=1226916126 qtime=1226916126
etime=1226916126 start=1226921658 owner=alicesgm at XXXX.ru
exec_host=n20.lcgwn.kiae/2 Resource_List.cput=48:00:00
Resource_List.neednodes=1 Resource_List.nodect=1 Resource_List.nodes=1
Resource_List.walltime=72:00:00 session=4123 end=1226922518
Exit_status=0 resources_used.cput=00:00:01 resources_used.mem=16216kb
resources_used.vmem=70092kb resources_used.walltime=00:-45:-53


Now I would be happy to ingore one or two and blame it on bad memory or
something but these are seen reported back from around 50 sites.

If no obvious suggestions as to why -'ve walltime might happen I will
get some more information
about versions etc, ...
Any ideas?



More information about the torqueusers mailing list