mpich-gm1.2.5..10 and PBSpro

"James Lowey" <[email protected]> Tue, 14 Oct 2003 16:38:47 -0700
Newsgroups gmane.network.myrinet.general
Message-ID <[email protected]>
I am having some problems getting mpich-gm and pbspro to work correctly
together.
If I submit a sample job using pbspro I get the following output:

Ignoring message from the MPI Id 0 (./cpi) !
Ignoring message from the MPI Id 0 (./cpi) !
Ignoring message from the MPI Id 0 (./cpi) !
Ignoring message from the MPI Id 0 (./cpi) !
Ignoring message from the MPI Id 0 (./cpi) !

This will go on for 1000's and 1000's of lines then
you will see some of these messages:

Received spurious abort message, keep listening...
Received spurious abort message, keep listening...
Received spurious abort message, keep listening...
Received spurious abort message, keep listening...
Received spurious abort message, keep listening...
Received spurious abort message, keep listening...

Until the job is killed.  The job will run successfully but it appears
that it outputs every
message that is there.  We came up with a workaround by modifying the
mpirun.ch_gm script 
setting warnings to 0 but when you submit a job via PBSpro the head mom
node shows a perl process from mpich-gm consuming a tremendous amount of
processor resources. So the "fix" only masks the problem, it does not
solve it.

System info:
myrinet driver version 1.6.4
redhat 7.3
pbspro 5.3
128 myrinet nodes

It seems like PBS is setting an environmental variable or something that
makes it so mpich-gm sends every message to stdout.  But PBS works fine
with MPICH-p4. If anybody has any ideas I would love to hear them!

Thanks, 

James Lowey
Sr. Unix Administrator
TGen,  The Translational Genomics Research Institute
400 N. Fifth Street,  Suite 1600
Phoenix, AZ 85004
http://www.tgen.org