[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] Condor G MPI - relevant RSL for processors per node



Hi,

I was trying to port a workflow to a new cluster where I have to specify in a corresponding pbs file that :ppn=4.

Since my workflow uses condorg, How do i specify the token corresponding to ppn=4 in condor submit file.?

my submit file initially was:

Executable = /panasas/scratch/sirish/ww3-test-dir/WANAFp08-UFL_20050827T0000_20050827T0000_20050901T0000_12gsr1-T486_Z-20070910-16483894c856b8-5fe7-11dc-885a-0011432efe34/ww3_shel
Universe = globus
Globusscheduler = cluster.loni.org/jobmanager-pbs
Transfer_Executable = false
Log = WW3.ERDF.12.200707181016.20070910-16483894c856b8-5fe7-11dc-885a-0011432efe34.workflow.log
Output = WW3.ERDF.12.200707181016.20070910-16483894c856b8-5fe7-11dc-885a-0011432efe34.ww3_shel.out
Error = WW3.ERDF.12.200707181016.20070910-16483894c856b8-5fe7-11dc-885a-0011432efe34.ww3_shel.err
Notification = never
remote_initialdir = /panasas/scratch/sirish/ww3-test-dir/WANAFp08-UFL_20050827T0000_20050827T0000_20050901T0000_12gsr1-T486_Z-20070910-16483894c856b8-5fe7-11dc-885a-0011432efe34
globusrsl = (jobType = mpi) (count = 4) (max_wall_time = 45) (queue = checkpt) (project = test_allocation03)
Queue

I have tried many options but I am getting the following error:

018 (1350.000.000) 09/10 17:18:59 Globus job submission failed!
    Reason: 48 the provided RSL could not be properly parsed
...
012 (1350.000.000) 09/10 17:18:59 Job was held.
        Globus error 48: the provided RSL could not be properly parsed
        Code 2 Subcode 48

It clearly shows that something is missing.

Does anybody have an idea of how to specify processors per node in a condor/condorg submit file?  Thank you for your support.

sirish


Tonight's top picks. What will you watch tonight? Preview the hottest shows on Yahoo! TV.