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

[Condor-users] MPICH2 and Condor Help!!!

Hi All!

I've installed MPICH2 and if I run the program with mpirun (not with condor)
they run perfectly, using the processors I need, etc,BUT If I use condor
whit this submit file:

universe = parallel
executable = mp2script
log = logfile
input = infile.$(NODE)
output = outfile.$(NODE)
error = errfile.$(NODE)
aguments = paralelo 128
machine_count = 3
should_transfer_files = yes
when_to_transfer_output = on_exit
transfer_input_files = paralelo

Can the dedicated scheduler and the Condor master be different?

The ScheddLog of the submitter says:
Negotiating for owner: DedicatedScheduler@xxxxxxxxxxxxxxxxxxxxxxxxx
12/1 20:23:33 (pid:3418) Out of requests - 1 reqs matched, 0 reqs idle
12/1 20:23:33 (pid:3418) Response problem from startd on <>
(match <>#1227875667#1).

And the StartLog of the executing machine says:
 DC_AUTHENTICATE: Unable to reconcile!
12/1 20:23:36 DaemonCore: PERMISSION DENIED for 442 via TCP from host
12/1 20:23:36 DC_AUTHENTICATE: Unable to reconcile!

Any ideas? Thanks,

Ing. Paula Marti­nez
ITU - Redes y Telecomunicaciones