[mpich-discuss] errors with MPICH2

Rajeev Thakur thakur at mcs.anl.gov
Wed Feb 18 14:35:30 CST 2009


Probably something with the networking configuration on some of the machines
that prevents processes from contacting each other. You need to run mpdcheck
between every pair of nodes, particularly the ones that show the error.
 
Rajeev


  _____  

From: mpich-discuss-bounces at mcs.anl.gov
[mailto:mpich-discuss-bounces at mcs.anl.gov] On Behalf Of
robert.mulloy at shell.com
Sent: Wednesday, February 18, 2009 1:50 PM
To: mpich-discuss at mcs.anl.gov
Subject: [mpich-discuss] errors with MPICH2




i've compiled MPICH2 and am getting frustrating errors. when i run mpdcheck
it works fine on any 1 or 2 nodes but when i go to more than 2 nodes i keep
getting errors (handle output 393, 384), failed handshake and its a random
node that causes the failure. 

has this happened to anyone else before? 

thanks 

rob 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/mpich-discuss/attachments/20090218/672f97c8/attachment.htm>


More information about the mpich-discuss mailing list