[OMPI users] How to get rid of OpenMPI warning: unable to find any relevant network interfaces
I am using OpenMPI 1.10.1 built with Intel compilers 16.0.3 on Centos 6.7. Our cluster has some cells where the nodes have infiniband and some cell with 10GB IP. I am hoping that I can build just one version of OpenMPI that can run on both infiniband and IP cells and not get any network-related warnings. I configured and built OpenMPI on a node with infiniband as follows: ./configure CXX=icpc CC=icc FC=ifort --prefix=/opt/openmpi/1.10.1-intel16.0 \ --enable-mpirun-prefix-by-default --with-threads --enable-mpi-thread-multiple \ --with-tm=/opt/torque/active --with-verbs=/usr --with-verbs-libdir=/usr/lib64 We are using OFED 3.2 that is part of the Centos distribution. When I run a program built with this version of OpenMPI and configured as described above on a cell with infiniband, the program runs without any warnings or errors. However, when I run on an IP-based cell, I get the following warning: -- A high-performance Open MPI point-to-point messaging module was unable to find any relevant network interfaces: Module: OpenFabrics (openib) Host: k07 Another transport will be used instead, although this may result in lower performance. — The program runs correctly but I am wondering if there is a way to get rid of the warning. I am aware of the MCA btl arguments that can be used with mpirun to specify ib or ip but I was hoping I didn’t have to burden my users with those details. Is there a way to configure OpenMPI to run on either IB or IP without warnings. Thank you, Susan Susan Schwarz Research Software Engineer Research Computing Dartmouth College Hanover, NH 03755 ___ users mailing list users@lists.open-mpi.org https://rfd.newmexicoconsortium.org/mailman/listinfo/users
Re: [OMPI users] Is gridengine integration broken in openmpi 2.0.2?
Will this be fixed in the 2.0.3 release? Thanks. Glenn On Mon, Feb 6, 2017 at 10:45 AM, Mark Dixon wrote: > On Mon, 6 Feb 2017, Mark Dixon wrote: > ... > >> Ah-ha! "-mca plm_rsh_agent foo" fixes it! >> >> Thanks very much - presumably I can stick that in the system-wide >> openmpi-mca-params.conf for now. >> > ... > > Except if I do that, it means running ompi outside of the SGE environment > no longer works :( > > Should I just revoke the following commit? > > Cheers, > > Mark > > commit d51c2af76b0c011177aca8e08a5a5fcf9f5e67db > Author: Jeff Squyres > Date: Tue Aug 16 06:58:20 2016 -0500 > > rsh: robustify the check for plm_rsh_agent default value > > Don't strcmp against the default value -- the default value may change > over time. Instead, check to see if the MCA var source is not > DEFAULT. > > Signed-off-by: Jeff Squyres > > (cherry picked from commit open-mpi/ompi@71ec5cfb436977ea > 9ad409ba634d27e6addf6fae) > > > ___ > users mailing list > users@lists.open-mpi.org > https://rfd.newmexicoconsortium.org/mailman/listinfo/users > ___ users mailing list users@lists.open-mpi.org https://rfd.newmexicoconsortium.org/mailman/listinfo/users
Re: [OMPI users] Is gridengine integration broken in openmpi 2.0.2?
Yes, we can get it fixed.
Ralph is unavailable this week; I don't know offhand what he meant by his prior
remarks. It's possible that
https://github.com/open-mpi/ompi/commit/71ec5cfb436977ea9ad409ba634d27e6addf6fae;
can you try changing the "!=" on line to be "=="? I.e., from
if (MCA_BASE_VAR_SOURCE_DEFAULT != source) {
to
if (MCA_BASE_VAR_SOURCE_DEFAULT == source) {
I filed https://github.com/open-mpi/ompi/issues/2947 to track the issue.
> On Feb 9, 2017, at 6:01 PM, Glenn Johnson wrote:
>
> Will this be fixed in the 2.0.3 release?
>
> Thanks.
>
>
> Glenn
>
> On Mon, Feb 6, 2017 at 10:45 AM, Mark Dixon wrote:
> On Mon, 6 Feb 2017, Mark Dixon wrote:
> ...
> Ah-ha! "-mca plm_rsh_agent foo" fixes it!
>
> Thanks very much - presumably I can stick that in the system-wide
> openmpi-mca-params.conf for now.
> ...
>
> Except if I do that, it means running ompi outside of the SGE environment no
> longer works :(
>
> Should I just revoke the following commit?
>
> Cheers,
>
> Mark
>
> commit d51c2af76b0c011177aca8e08a5a5fcf9f5e67db
> Author: Jeff Squyres
> Date: Tue Aug 16 06:58:20 2016 -0500
>
> rsh: robustify the check for plm_rsh_agent default value
>
> Don't strcmp against the default value -- the default value may change
> over time. Instead, check to see if the MCA var source is not
> DEFAULT.
>
> Signed-off-by: Jeff Squyres
>
> (cherry picked from commit
> open-mpi/ompi@71ec5cfb436977ea9ad409ba634d27e6addf6fae)
>
>
> ___
> users mailing list
> users@lists.open-mpi.org
> https://rfd.newmexicoconsortium.org/mailman/listinfo/users
>
> ___
> users mailing list
> users@lists.open-mpi.org
> https://rfd.newmexicoconsortium.org/mailman/listinfo/users
--
Jeff Squyres
jsquy...@cisco.com
___
users mailing list
users@lists.open-mpi.org
https://rfd.newmexicoconsortium.org/mailman/listinfo/users
Re: [OMPI users] How to get rid of OpenMPI warning: unable to find any relevant network interfaces
Susan -- Try setting --mca btl_base_warn_component_unused 0 That should make the warning go away (shame on us for not putting that in the warning message itself -- doh!). If that works for you, you can put "btl_base_warn_component_used = 0" in $prefix/etc/openmpi-mca-params.conf (i.e., your side-wide configuration). > On Feb 9, 2017, at 3:30 PM, Susan A. Schwarz > wrote: > > I am using OpenMPI 1.10.1 built with Intel compilers 16.0.3 on Centos 6.7. > Our cluster has some cells where the nodes have infiniband and some cell with > 10GB IP. I am hoping that I can build just one version of OpenMPI that can > run on both infiniband and IP cells and not get any network-related warnings. > I configured and built OpenMPI on a node with infiniband as follows: > > ./configure CXX=icpc CC=icc FC=ifort --prefix=/opt/openmpi/1.10.1-intel16.0 \ > --enable-mpirun-prefix-by-default --with-threads --enable-mpi-thread-multiple > \ > --with-tm=/opt/torque/active --with-verbs=/usr --with-verbs-libdir=/usr/lib64 > > We are using OFED 3.2 that is part of the Centos distribution. When I run a > program built with this version of OpenMPI and configured as described above > on a cell with infiniband, the program runs without any warnings or errors. > However, when I run on an IP-based cell, I get the following warning: > > -- > A high-performance Open MPI point-to-point messaging module > was unable to find any relevant network interfaces: > Module: OpenFabrics (openib) > Host: k07 > > Another transport will be used instead, although this may result in > lower performance. > — > The program runs correctly but I am wondering if there is a way to get rid of > the warning. I am aware of the MCA btl arguments that can be used with > mpirun to specify ib or ip but I was hoping I didn’t have to burden my users > with those details. Is there a way to configure OpenMPI to run on either IB > or IP without warnings. > > Thank you, > Susan > > Susan Schwarz > Research Software Engineer > Research Computing > Dartmouth College > Hanover, NH 03755 > > ___ > users mailing list > users@lists.open-mpi.org > https://rfd.newmexicoconsortium.org/mailman/listinfo/users -- Jeff Squyres jsquy...@cisco.com ___ users mailing list users@lists.open-mpi.org https://rfd.newmexicoconsortium.org/mailman/listinfo/users
