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

[Condor-users] Checkpoint and Migration

Hi, when I submit a job to the condor pool it go to be executed in the machine A, because it state is Unclaimed  and activity is Idle... if the machine A user becomes and change it state to Owner, the condor pool stop the job and migrate it to the machine B... however the job is restarted on machine B instead to continue.... my submit file contents is:

universe        = vanilla
executable      = zr2o4ti2o4.sh
output          = zr2o4ti2o4.sh.out
error           = zr2o4ti2o4.sh.err
log             = zr2o4ti2o4.sh.log
should_transfer_files = IF_NEEDED
when_to_transfer_output = ON_EXIT

How can I configure it to checkpoint works?

Thanks very much!!!