Cancelled job in the middle of execution

[Nikhil-Shenoy]

When we submit the pilot and add it to the unit manager, the following code gets executed:

pdesc = rp.ComputePilotDescription ()
pdesc.resource = "yarn.aws-vm"  # NOTE: This is a "label", not a hostname
pdesc.runtime  = 30 # minutes
pdesc.cores    = 1
pdesc.cleanup  = False
# submit the pilot.
print "Submitting Compute Pilot to Pilot Manager ..."
pilot = pmgr.submit_pilots(pdesc)
umgr.add_pilots(pilot)

The script waits indefinitely at this point and the job had to be manually cancelled

Session info: Session id: rp.session.ip-10-99-194-113.ec2.internal.test099.016751.0001 Pilot Manager: {'uid': 'pmgr.0001'}

Username = test099

[andre-merzky]

Did the call to umgr.add_pilots() return, or did that one hang? If it did return, what was the next call, ie. which specific call did not return?

Thanks!

[iparask]

Andre,

can you point me in which log file to look. I am checking right now the log files in the run Nikhil points.

[andre-merzky]

I am not sure if that is visible in the logs. The statement above was:

The script waits indefinitely at this point

and I am not sure I understand what this point refers to... Did that last call hang, or did it hang afterwards? Thanks!