Dear all,
I was running the simulation on parallel computations within R. The
procedure uses the "LAM/MPI" to communicate between nodes. However, the
simulation was terminated unexpectely and the error message was "MPI process
rank 0 (node 0) caught a SIGBUS error".
When I type "lamclean -v", it returned the message "It seems that there is
no lamd running on the host head1002.local". I tried to reboot using command
"lamboot -v .hostfile". However MPI failed to boot. I wonder what happened
and how I can fix this problem.
Thanks,
Iris
|
