[mvapich-discuss] Fatal Error in MPI_Init
Lockwood, Glenn
glock at sdsc.edu
Mon Nov 25 12:19:36 EST 2013
The operative error is "Connection refused," and I assume you used mpirun_rsh, so it sounds like you don't have ssh configured properly between your two nodes.
Glenn
On Nov 24, 2013, at 12:38 AM, Sunny Soung <loesprite at gmail.com<mailto:loesprite at gmail.com>> wrote:
Hi there,
I'm new with MVAPICH2. I built a cluster with 2 nodes. When I run a distributed parallel task, I got the error message below.
----------------------------------------------------------------------------------------------------------------------------------------------------
[cli_0]: aborting job:
Fatal error in MPI_Init:
Other MPI error
[cli_1]: aborting job:
Fatal error in MPI_Init:
Other MPI error
[master:mpispawn_0][child_handler] MPI process (rank: 0, pid: 11242) exited with status 1
[node1:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 6. MPI process died?
[node1:mpispawn_1][read_size] Unexpected End-Of-File on file descriptor 6. MPI process died?
[node1:mpispawn_1][handle_mt_peer] Error while reading PMI socket. MPI process died?
[node1:mpispawn_1][child_handler] MPI process (rank: 1, pid: 10156) exited with status 1
[node1:mpispawn_1][report_error] connect() failed: Connection refused (111)
----------------------------------------------------------------------------------------------------------------------------------------------------
Please help me to find the cause.
Thanks in advance.
Sunny
_______________________________________________
mvapich-discuss mailing list
mvapich-discuss at cse.ohio-state.edu<mailto:mvapich-discuss at cse.ohio-state.edu>
http://mailman.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20131125/037a1a31/attachment.html>
More information about the mvapich-discuss
mailing list