It's true, with this jdl
Type = "Job";
JobType = "normal";
VirtualOrganisation= "biomed";
Executable = "/bin/hostname";
StdOutput = "salida.out";
StdError = "error.err";
OutputSandbox = {"error.err","salida.out"};
Requirements = other.GlueCEUniqueID ==
"mallarme.cnb.uam.es:2119/jobmanager-lcgpbs-biomed";
the job seems to be "hanged", with edg-job-status the job is all the
time "scheduled".
In the CE
[dteam001@mallarme pbs]$ qstat
Job id Name User Time Use S Queue
---------------- ---------------- ---------------- -------- - -----
635.mallarme STDIN biomed002 0 R biomed
636.mallarme STDIN biomed002 0 R biomed
What are the most interesting PBS logs in /var/spool/pbs to check this
errors?
Thank you advanced.
Burke, S (Stephen) wrote:
>LHC Computer Grid - Rollout
>
>
>>[mailto:[log in to unmask]] On Behalf Of David
>>Garcia Aristegui said:
>>Requirements = other.GlueCEUniqueID == "mallarme.cnb.uam.es";
>>
>>
>
>The CEUniqueID is the full contact string, i.e.
>mallarme.cnb.uam.es:2119/jobmanager-lcgpbs-biomed.
>
>
>
>>[david@villon david]$ globus-job-run
>>mallarme.cnb.uam.es/jobmanager-lcgpbs /bin/hostname
>>
>>Nothing, there is no output.
>>
>>
>
>Do you see anything in the PBS logs when you do that?
>
>Stephen
>
>
>
>
|