[mvapich-discuss] mvapich-0.9.8 Bus error

[Rene Salmon]

Hi list,

I am very new to mvapich and  I am having problems getting it to run.  I
compiled mvapich-0.9.8 on our cluster which runs the Voltaire version of the
OFED stack.  To install I simply ran the "make.mvapich.gen2" script after
pointing it to the /usr/local/ofed dir for IBHOME.

I can run a 2 or 4 CPU job on two nodes just fine.  The problem happens when
I try to run a 6 or greater CPU job.  I get a "Bus error" message. Here it
is running on 2 nodes with 4 CPUs per node.

rsalmon at login-02-01 177> mpirun_rsh -np 8 -hostfile nodelist.txt ./a.out
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
/usr/X11R6/bin/xauth:  error in locking authority file
/u00/rsalmon/.Xauthority
Bus error
Bus error
Bus error
Bus error
Bus error
Bus error
Bus error
Bus error


rsalmon at login-02-01 178> cat nodelist.txt
compute-01-01-ib
compute-01-01-ib
compute-01-01-ib
compute-01-01-ib
compute-01-02-ib
compute-01-02-ib
compute-01-02-ib
compute-01-02-ib





Any ideas as to what might be wrong?
Thank you 
Rene







-- 
        Rene Salmon
        Tulane University
        Center for Computational Science
        http://www.ccs.tulane.edu
        rsalmon at tulane.edu
        Tel 504-862-8393
        Fax 504-862-8392