Open MPI logo

Open MPI User's Mailing List Archives

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

Subject: Re: [OMPI users] problem with opal_net_private_ipv4
From: Fernando Lemos (fernandotcl_at_[hidden])
Date: 2010-03-23 11:28:54

On Tue, Mar 23, 2010 at 10:25 AM, Nicolas Niclausse
<Nicolas.Niclausse_at_[hidden]> wrote:
> Hello,
> I'm trying to run openmpi (1.4.1) on two clusters; on each cluster, several
> interfaces are private;
> on cluster1, nodes have 3 interfaces, and only is visible
> from cluster2.
> chicon-3
> eth0     inet addr:  Bcast:  Mask:
> eth1     inet addr:  Bcast:  Mask:
> myri0    inet addr:  Bcast:  Mask:
> on cluster2, nodes have 3 interfaces, and only is visible
> from cluster1
> netgdx-8
> eth0  inet addr:  Bcast:  Mask:
> eth1  inet addr:  Bcast:  Mask:
> eth2  inet addr:  Bcast:  Mask:
> so i'm using this to declare all the other networks as private:
> mpirun -machinefile ~/gridnodes  --mca opal_net_private_ipv4
> "\;\;\;"
> ./alltoall
> but this doesn't work:

Have you tried -mca btl_tcp_if_include/exclude?

> Why openmpi tries to connect different private networks, given that
> "public" networks exists ? is it a bug or am i missing something ?

>From what I've seen, I believe OpenMPI tries to find the fastest route
to the nodes. In some cases it's trivial to sort that out, in other
cases you might need to give it some hints.