Open MPI logo

Open MPI User's Mailing List Archives

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

Subject: Re: [OMPI users] Segfaults w/ both 1.4 and 1.5 on CentOS 6.2/SGE
From: Joshua Baker-LePain (jlb17_at_[hidden])
Date: 2012-03-13 22:56:48

On Tue, 13 Mar 2012 at 6:05pm, Ralph Castain wrote

> I started playing with this configure line on my Centos6 machine, and
> I'd suggest a couple of things:
> 1. drop the --with-libltdl=external ==> not a good idea
> 2. drop --with-esmtp ==> useless unless you really want pager messages notifying you of problems
> 3. drop --enable-mpi-threads for now
> I'm continuing to play with it, but thought I'd pass those along.

After my first custom build of 1.4.5 didn't work, I built it again using
an utterly minimal configure line:

./configure --prefix=$HOME/ompi-1.4.5 --with-sge

Runs with this library still failed, but the backtrace did change

#0 0x00002b7bbc8681d0 in ?? ()
#1 <signal handler called>
#2 0x00002b7bbd2b8f6c in mca_btl_sm_component_progress ()
    from /netapp/sali/jlb/ompi-1.4.5/lib/openmpi/
#3 0x00002b7bb9b2feda in opal_progress ()
    from /netapp/sali/jlb/ompi-1.4.5/lib/
#4 0x00002b7bba9a98d5 in barrier ()
    from /netapp/sali/jlb/ompi-1.4.5/lib/openmpi/
#5 0x00002b7bb965d426 in ompi_mpi_init ()
    from /netapp/sali/jlb/ompi-1.4.5/lib/
#6 0x00002b7bb967cba0 in PMPI_Init ()
    from /netapp/sali/jlb/ompi-1.4.5/lib/
#7 0x0000000000400826 in main (argc=1, argv=0x7fff93634788)
     at mpihello-long.c:11

Joshua Baker-LePain
QB3 Shared Cluster Sysadmin