Open MPI logo

Open MPI User's Mailing List Archives

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

From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2006-02-14 17:43:03


Ok, we've finally migrated over all the I/O forwarding fixes to the
v1.0 branch. They've been on the trunk for a little while, but we
have certain procedures that have to be followed to get things
committed on the release branch and it took a little time -- sorry
about that.

Can you try a nightly tarball from tonight -- either the trunk or a
v1.0 tarball and let us know how it goes?

On Feb 13, 2006, at 9:46 AM, Andy Vierstraete wrote:

> Hi Peter,
>
> I tried first with 2 nodes, but is was the same problem, it hang.
> Then I tried with 1 node, and copied that output in my previous
> mail. When I checked the process status (with 2 nodes), one of the
> migrate processes was using 100 %, the other was sleeping. But
> after 15 minutes, still no output change.
>
> Andy
>
> Peter Beerli wrote:
>> Dear Andy, you wrote that with openmpi:
>>> avierstr_at_muscorum:~> mpiexec --hostfile hostfile -np 1 migrate-n
>> it does not work, but with lam-mpi
>>> avierstr_at_muscorum:~/thomas> mpiexec -np 2 migrate-n
>> you started openmpi on only _one_ node, migrate needs at least
>> _two_ nodes to run (as you did in lam-mpi) migrate actually aborts
>> when running on only one node, it should show an error message so,
>> like this zork>which mpirun /usr/local/openmpi/bin/mpirun
>> zork>mpirun -machinefile ~/onehost -np 1 migrate-n migrate-n
>> ============================================= MIGRATION RATE AND
>> POPULATION SIZE ESTIMATION using Markov Chain Monte Carlo
>> simulation ============================================= compiled
>> for a PARALLEL COMPUTER ARCHITECTURE Version debug 2.1.3 [x]
>> Program started at Mon Feb 13 09:03:45 2006 Reading N ... Reading
>> S ... In file main.c on line 697 This program was compiled to use
>> a parallel computer and you tried to run it on only a single node.
>> This will not work because it uses a "single_master-many_worker"
>> architecture and needs at least TWO nodes Peter
>> _______________________________________________ users mailing list
>> users_at_[hidden] http://www.open-mpi.org/mailman/listinfo.cgi/users
>
> --
> *********************************************************************
> * Youth is a wonderful thing. What a crime to waste it on children.
> * * (George Bernard Shaw) *
> *********************************************************************
> Andy Vierstraete Department of Biology University of Ghent K. L.
> Ledeganckstraat 35 B-9000 Gent Belgium phone : 09-264.52.66 fax :
> 09-264.87.93 http://allserv.UGent.be/~avierstr/
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users

-- 
{+} Jeff Squyres
{+} The Open MPI Project
{+} http://www.open-mpi.org/