[Mauiusers] Jobs stuck in deferred state and unstartable ...

[Richard Walsh]

All,

I have a list of jobs that were submitted after the PBS server
entered a problem state related to jobs/processes that it was
managing being suspended (state T on ps).  I have cleared
this up and now find that I still cannot start these jobs which were:

Holds:    Defer  (hold reason:  RMFailure)

The RM failure message for the un-runnable jobs is:

job is deferred.  Reason:  RMFailure
(cannot start job - RM failure, rc: 15044, msg: 'Resource temporarily 
unavailable')

Resetting the hold state with "releasehold -a jobid" does not
work.  All attempts at a forced "runjob -x|-f jobid" indicate that
deferred state jobs cannot be started:

job '15671' is in expected state 'Deferred' (expected state must be idle)

Job being submitted at present are starting and running without a
problem.  I have restarted both PBS and Maui twice.  I plan to
kill the jobs and have them resubmitte, but would like to have a
better solution for future reference.

Any suggestions?

rbw