Hello,
I can run the c, c++, R application under CONDOR successfully. But i am facing problems on running mpi under condor.
For configuration the DEDICATED SCHEDULER & DEDICATED RESOURCE, i used this reference:
https://lists.cs.wisc.edu/archive/condor-users/2007-February/msg00001.shtml
For your information:
* My central manager and nodes are ubuntu 10.04.
* $CondorVersion: 7.2.4 Apr 11 2010 $
* $CondorPlatform: I386-LINUX_DEBIAN_UNKNOWN $
* MPICH2 Version: 1.2.1p1
* I installed CONDOR using UBUNTU SOFTWARE CENTER.
* I have enable ssh passwordless.
* My condor_status:
OpSys Arch State Activity LoadAv Mem
LINUX INTEL
Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.000 997
LINUX INTEL Unclaimed Idle 0.230 755
LINUX INTEL Unclaimed Idle 0.000 755
LINUX INTEL Unclaimed
Idle 0.000 755
LINUX INTEL Unclaimed Idle 0.000 755
* This is my submission file:
executable = mp2script
arguments = hello_mpi
machine_count = 8
universe = parallel
output = out.$(NODE)
error = err.$(NODE)
log = log
should_transfer_files = yes
when_to_transfer_output = on_exit
transfer_input_files = hello_mpi
queue
* On mp2script file, i just edit this:
######################################################
# Set this to the bin directory of MPICH installation
MPDIR=/usr/bin
PATH=$MPDIR:.:$PATH
export PATH
######################################################
* When i run the job, i get this result from condor_q -better-analyze:
10 match
but reject the job for unknown reasons
Do you know how to solve this problem?
Thank you so much before :)