On 09/07/2012 03:49 PM, David Rebatto wrote:
> Hi!
>
> On 09/07/2012 03:33 PM, Sean Crosby wrote:
>> Things look okay authentication wise: (agh5 is our CE, agc55 is a
>> worker node)
>>
>> [pilatl09@agc55 ~]$ scp -r
>> [log in to unmask]:/opt/glite/var/cream_sandbox/atlaspil/_C_CA_O_Grid_OU_triumf_ca_CN_Asoka_De_Silva_GC1_atlas_Role_pilot_Capability_NULL_pilatl09
>> .
>> runpilot3-wrapper.sh 100% 8940 8.7KB/s
>> 00:00
>> StandardOutput 100% 70 0.1KB/s
>> 00:00
>> CREAM131815721_jobWrapper.sh 100% 24KB 23.5KB/s
>> 00:00
>> CREAM676695673_jobWrapper.sh 100% 24KB 23.5KB/s
>> 00:00
>> CREAM670609167_jobWrapper.sh 100% 23KB 23.5KB/s
>> 00:00
>> CREAM675568430_jobWrapper.sh 100% 24KB 23.5KB/s
>> 00:00
>> runpilot3-wrapper.sh 100% 8940 8.7KB/s
>> 00:00
>>
>> It could be a gridmap problem (CE has had a yaim reconfiguration, and
>> same with all the worker nodes, but not the Torque server). What I
>> don't get is that the file that the worker node tries to transfer is
>> "stageout=err_cream_772549819_StandardOutput" - this file doesn't
>> exist! The file that does exist is err_cream_772549819_StandardOutput
>> (i.e. no "stageout=")
>
> This sounds like a wrong format for the -W option in qsub command is
> being used.
> You could try to test different values for the
> "blah_torque_multiple_staging_directive_bug" in blah.conf.
>
> Cheers,
> David
P.S.: here's the relevant bug and fix explanation:
https://savannah.cern.ch/bugs/?89527
--
David Rebatto
I.N.F.N. - Sezione di Milano
Via Celoria, 16 - 20133 Milano ITALY
tel: +39 02503.17623 e-mail: [log in to unmask]
URL: http://www.mi.infn.it/~rebatto
"There are 10 kinds of people in the world:
those who understand binary and those who don't..."
|