Hi,

Just an update on this. I've managed to build OpenMPI 4.0.1 using

--with-pmix=/lustre/z04/pmix --width-slurm
-with-libevent=/lustre/z04/pmix/libevent/

Unfortunately I keep getting the same warning when using srun (i.e. srun
my_ompi_app.x):

The application appears to have been direct launched using "srun",
but OMPI was not built with SLURM's PMI support and therefore cannot
execute. There are several options for building PMI support under
SLURM, depending upon the SLURM version you are using:

  version 16.05 or later: you can use SLURM's PMIx support. This
  requires that you configure and build SLURM --with-pmix.

  Versions earlier than 16.05: you must use either SLURM's PMI-1 or
  PMI-2 support. SLURM builds PMI-1 by default, or you can manually
  install PMI-2. You must then build Open MPI using --with-pmi pointing
  to the SLURM PMI library location.

Please configure as appropriate and try again.

We use slurm 20.02.3 so configuring --with-pmix it is the correct way
according to this message. My worry is if we need to use the same PMIx
library that was used to build slurm, because that one has not been made
available.

Regards,
Luis



On 05/08/2020 16:10, Luis Cebamanos via devel wrote:
> Hi Ralph,
>
> Thanks for pointing me to this. I've done that and although configure
> does not report any errors, make won't build with following errors
>
> Making all in mca/pmix/ext3x
> make[2]: Entering directory '/lustre/z04/openmpi-4.0.1/opal/mca/pmix/ext3x'
>   CC       mca_pmix_ext3x_la-ext3x_local.lo
>   CC       mca_pmix_ext3x_la-ext3x.lo
>   CC       mca_pmix_ext3x_la-ext3x_client.lo
>   CC       mca_pmix_ext3x_la-ext3x_component.lo
> In file included from ext3x_local.c:21:
> ext3x.h:35:10: fatal error: pmix_server.h: No such file or directory
>  #include "pmix_server.h"
>           ^~~~~~~~~~~~~~~
>
> My configure options are as follow
>
> ./configure --prefix=/lustre/z04/ompi_slurm 
> --with-pmix=/lustre/z04/pmix --with-pmi=/lustre/z04/pmix --with-slurm
> --with-cuda=/lustre/sw/nvidia/hpcsdk/Linux_x86_64/cuda/10.1/include/
> --with-libevent=/lustre/z04/pmix/libevent/
>
> and I can see the that header file in the pmix built:
> $ls /lustre/z04/pmix/include/
> pmi2.h  pmix_common.h  pmix.h         pmix_server.h  pmix_version.h
> pmi.h   pmix_extend.h  pmix_rename.h  pmix_tool.h
>
> What am I missing?
>
> Cheers,
> Luis
> On 05/08/2020 15:21, Ralph Castain via devel wrote:
>> For OMPI, I would recommend installing PMIx: 
>> https://github.com/openpmix/openpmix/releases/tag/v3.1.5
>>
>>
>>> On Aug 5, 2020, at 12:40 AM, Luis Cebamanos via devel 
>>> <devel@lists.open-mpi.org> wrote:
>>>
>>> Hi all,
>>>
>>> We are trying to install OpenMPI with Slurm support on a recently
>>> upgraded system. Unfortunately libpmi, libpmi2 or limpix don't seem to
>>> be available. Could I install these libraries myself to build OpenMPI
>>> and run it with Slurm? If so, which one would be needed? I guess these
>>> libraries should have been installed and made available whith the Slurm
>>> installation, but unfortunately they had not...
>>>
>>> checking for pmi.h... yes
>>> checking for libpmi in /usr/include/slurm/lib... checking for libpmi in
>>> /usr/include/slurm/lib64... not found
>>> checking for pmi2.h in /usr/include/slurm... found
>>> checking pmi2.h usability... yes
>>> checking pmi2.h presence... yes
>>> checking for pmi2.h... yes
>>> checking for libpmi2 in /usr/include/slurm/lib... checking for libpmi2
>>> in /usr/include/slurm/lib64... not found
>>> checking for pmix.h in /usr/include/slurm... not found
>>> checking for pmix.h in /usr/include/slurm/include... not found
>>> checking can PMI support be built... no
>>> configure: WARNING: PMI support requested (via --with-pmi) but neither
>>> pmi.h,
>>> configure: WARNING: pmi2.h or pmix.h were found under locations:
>>> configure: WARNING:     /usr/include/slurm
>>> configure: WARNING:     /usr/include/slurm/slurm
>>> configure: WARNING: Specified path: /usr/include/slurm
>>> configure: WARNING: OR neither libpmi, libpmi2, or libpmix were found under:
>>> configure: WARNING:     /lib
>>> configure: WARNING:     /lib64
>>>
>>>
>>> Regards,
>>> Luis
>>>
>>> The University of Edinburgh is a charitable body, registered in Scotland, 
>>> with registration number SC005336.
>>> <pEpkey.asc>

-- 
~~~~~~~~~~~~~~~~~~~~~~~~~ | EPCC | ~~~~~~~~~~~~~~~~~~~~~~~~~

Luis Cebamanos, HPC Applications Consultant
Email: l.cebama...@epcc.ed.ac.uk Phone: +44 (0) 131 651 3479                   
http://www.epcc.ed.ac.uk/   
The Bayes Centre, 47 Potterrow, Edinburgh UK
EH8 9BT

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Attachment: pEpkey.asc
Description: application/pgp-keys

Reply via email to