[MPICH] Program running with mpich2 stops

[Yasmine Chebaro]

Hi all,

I am using mpich2 on a 64Bit Linux x86_64 machines, running in parrallel.
The program running with mpich2 is a mix of Fortran 90 and 77, with the
Intel Fortran Compiler.
I am actually having problems when I run my program with mpich2, in fact
after running for a couple of hours the program stops and the error message
is (ignore the four first lines they're the  "normal" output of my program)


         118   31.0400010000000
         119   12.0100000000000
         120   16.0000000000000
         121   16.0000000000000
place holderplace holder
Image              PC                Routine            Line        Source
simulateur         0000000000E3F3CE  Unknown               Unknown  Unknown
simulateur         0000000000E3E5CA  Unknown               Unknown  Unknown
simulateur         0000000000DF9F62  Unknown               Unknown  Unknown
simulateur         0000000000DC99BA  Unknown               Unknown  Unknown
simulateur         0000000000DC8F29  Unknown               Unknown  Unknown
simulateur         0000000000DDAB2B  Unknown               Unknown  Unknown
simulateur         0000000000438804  Unknown               Unknown  Unknown
simulateur         0000000000457C86  Unknown               Unknown  Unknown
simulateur         000000000045462D  Unknown               Unknown  Unknown
simulateur         0000000000411962  Unknown               Unknown  Unknown
libc.so.6          00002AE4FD3C34CA  Unknown               Unknown  Unknown
simulateur         00000000004118AA  Unknown               Unknown  Unknown
rank 1 in job 1  fargeau_44741   caused collective abort of all ranks
  exit status of rank 1: return code 104

I really don't know what to do know. I thought it was due to some stack
overflow with the Intel Compiler (which happened to me before), but even
with the option that solves the stack overflow problem, it doesnt work.
If you have any idea why is this happening.
Thanks in advance.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20070924/42b4e0fc/attachment.htm>