Hi,
okay... /var/glite/workload_manager/ismdump.fl is filled now
When doing a glite-job-list-match we get some resources e.g.:
deimos.ehv.campus.philips.com:2119/blah-pbs-phicos
mu6.matrix.sara.nl:2119/jobmanager-pbs-medium
mu6.matrix.sara.nl:2119/jobmanager-pbs-short
When we submit the job to mu6 the job completes succesfully.
When submitting to deimos, we see (on WMSLB):
MatchLog: 6/8 10:29:20 Rejected 1016.0 [log in to unmask] <194.171.252.154:22897>: no match found
NegotiatorLog: 6/8 10:29:20 Rejected 1016.0 [log in to unmask] <194.171.252.154:22897>: no match found
SchedLog:6/8 10:29:42 (pid:9904) Job 1016.0 put on hold: The job attribute PeriodicHold expression 'Matched =!= TRUE && CurrentTime > QDate + 900' evaluated to TRUE
So, deimos rejects the job, while the info systems thinks it is a match. Where should we look to find the problem?
Note that we are trying to get a gliteCE TORQUE to work after upgrading from lcg-CE_torque.
Grtz,
Serge
> -----Original Message-----
> From: LHC Computer Grid - Rollout
> [mailto:[log in to unmask]] On Behalf Of Burke,
> S (Stephen)
> Sent: Wednesday 6 June 2007 16:30
> To: [log in to unmask]
> Subject: Re: [LCG-ROLLOUT] WMSLB can't match any job to any CE
>
> LHC Computer Grid - Rollout
> > [mailto:[log in to unmask]] On Behalf Of
> > Vrijaldenhoven, Serge saidL
> > on WMSLB (which is also the BDII):
> > LCG_GFAL_INFOSYS=moon.ehv.campus.philips.com:2170
>
> I can't connect to that, is it behind a firewall?
>
> > Somehow /var/glite/workload_manager/ismdump.fl is very empty
> > on our system...
>
> Well, that would explain why the list-match doesn't work ...
>
> Stephen
>
|