Re: MPICH-GM problem: cannot run multiple executables in one job

Shubin Liu wrote:
if you put multiple copies in one script file ... and run it interactively, the job will hang after the first  executable is
done, with the output like this: ... Received spurious abort message, keep listening... \endquote

When running the standard MPI test on a 700 MHz Pentium III using the
latest MPICH-GM, many tests failed until I inserted
   --gm-wait 1
into mpich-1.2.5..9/bin/mpirun.ch_gm by changing the line to
 FLAGS="$GMPI_FLAGS --gm-wait 1 $MPICH_FLAGS"

On a dual-processor board with AMD XMP 2200+ the mpich tests ran fine
without adding --gm-wait.

I do not know if "--gm-wait 1" was really the solution, there may have
been other factors such as the load of the nodes during the various
times that I ran the tests, but it may be worth a try.
   The option "--gm-wait n" is an option for mpirun when ch_gm is
being used, it is not necessary to change the mpirun.ch_gm.  I changed
the file mpirun.ch_gm because I was running the built-in tests of mpich.
Alan Scheinine