I'm working AMD Opterons (Linux 64 bit) running RedHat Advanced Server
3.
This Linux distribution has LAM 6.5.9 in its default system commands
(/usr/bin).
I'm trying to use the LAM in /usr/bin area (version 6.5.9) instead of my
version which is version 6.3.2. I'm running into the following error
MPI_Init: LAM error: Unknown error 471
------------------------------------------------------------------------
-----
It seems that [at least] one of processes that was started with mpirun
did not invoke MPI_INIT before quitting (it is possible that more than
one process did not invoke MPI_INIT -- mpirun was only notified of the
first one, which was on node n0).
mpirun can *only* be used with MPI programs (i.e., programs that
invoke MPI_INIT and MPI_FINALIZE). You can use the "lamexec" program
to run non-MPI programs over the lambooted nodes.
I can run my test script with my version (6.3.2) but is running into
the problem when I use version 6.5.9 in /usr/bin. My application does
call MPI_Init() in parent and each of the child processes? I tried
running lamexec but it doesn't accept -c2c option.
Any help is appreciated greatly!!
YoungHui Amend
|
