I set up a cluster of two nodes by installi LAM MPI
7.1.1. Both of the nodes have the same h/w
configuration.
I used the command
>mpirun C -ssi rpi tcp chapter_10
and the error I got was this:
-----------------------------------------------------------------------------
It seems that [at least] one of the processes that was
started with
mpirun chose a different RPI than its peers. For
example, at least
the following two processes mismatched in their RPI
selections:
MPI_COMM_WORLD rank 1: tcp (v7.0.0)
MPI_COMM_WORLD rank 0: tcp (v7.1.0)
All MPI processes must choose the same RPI module and
version when
they start. Check your SSI settings and/or the local
environment
variables on each node.
-----------------------------------------------------------------------------
The laminfo command shows same version of the tcp rpi
used on the two nodes, which is 7.1.1.
What could be the probable mismatch? how can i set it
right?
I executed the hello application from examples, it
works fine. But the other applications do not work. I
also tried all the rpi modules (lamd, crtcp etc.) but
results in same kind of error.
Please help.
Thanx in advance
pictguys
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
|
