Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Open MPI User's mailing list

Subject: [OMPI users] Centos Open MPI with nfs problem
From: Kristian Medri (kmedri_at_[hidden])
Date: 2011-02-04 17:47:05


As long as the processes are on a single workstation I am having success.
Running processes across several workstation works as long as there is no
output to a file. When there is output to a file the CPUs seem to stall at
100% and provide the error:

[machine.1.com][[37707,1],2][btl_tcp_endpoint.c:638:mca_btl_tcp_endpoint_com
plete_connect] connect() to 192.168.122.1 failed: Connection
refused (111)

With exact machine configuration details, commands, and output at:

http://www.doe.carleton.ca/~kmedri/research/centosmeepinstall.html

Any hints on what to look for to get the secondary machines to behave?

I can provide additional details such as responses to 'rpcinfo -p', 'service
___ status' should the above link not be clear enough.

Sincere thanks!