From: Blas Pedro Uberuaga <buber@u.washington.edu>
Newsgroups: comp.parallel.mpi
Subject: Problem with MPICH and starting processes
Date: Tue, 06 Apr 1999 08:46:50 -0400
Organization: University of Washington
Message-Id: <370A023A.28A86802@u.washington.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Xref: ukc comp.parallel.mpi:4891


Hi All,

I am running MPICH 1.1.2 on a cluster of IBM 43P-260s.  When I start a
job, I often get a time out error causing me to cancel the job and
restart (this is interactively).  The exact error I get is :

net_recv_timeout failed for fd = 8
p6_13266: 258:  p4_error: net_recv_timeout read, errno = : 25

I'm assuming machines aren't responding in some default time.  Can I
increase that time somehow?  Thanks.