On Wed, 17 Jul 2002, VINOD wrote:
> I have run 'mpirun -np 4 xhpl', it gives error message
>
> bash: /HPL/hpl/bin/Linux_PI/xhpl : No such a file or directory.
>
> and after 5 minutes it gives error message
>
> P0_7037: P4_error: Timeout in making connection to remote process on
>oscarnode1.cluster: 0
> bm_list_7036: p4_error : interrupt SIGINT :2
>
> /opt/mpich-1.2.1/bin/mpirun: line 1: 7035 Brocken pipe
> /HPL/hpl/bin/Linux_4/xhpl -P4Pg /home/user1/PI6916 -P4wd
>/HPL/hpl/bin/Linux_4
The problem is that the directory /HPL (and all the directories below it)
is not available on all nodes. So mpirun is able to start it on the local
node, but then fails trying to start it up on the remote nodes, and the
local node eventually times out waiting for the remote nodes to contact it
(probably during MPI_INIT).
I'd suggest building your program under your $HOME directory (which should
be under /home, which is NFS mounted on all nodes).
{+} Jeff Squyres
{+} [EMAIL PROTECTED]
{+} http://www.lam-mpi.org/
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users
