[slurm-users] slurm-wlm package OpenMPI PMIx implementation

[Avery Grieve]

Hi Chris,

Thank you for the offer. Here's some quick information on my system:

All nodes on Debian 10 (armbian buster converted to DietPi v6.33.3).
sinfo --version: slurm-wlm 18.08.5-2

With MpiDefault=pmix I get the following srun errors:
srun: error: Couldn't find the specified plugin name for mpi/pmix looking
at all files
srun: error: cannot find mpi plugin for mpi/pmix
srun: error: cannot create mpi context for mpi/pmix
srun: error: invalid MPI type 'pmix', --mpi=list for acceptable types

With MpiDefault=none
I get OpenMPI yelling at me and giving me two options, only one relevant to
the version of Slurm I'm running:
version 16.05 or later: you can use SLURM's PMIx support. This
  requires that you configure and build SLURM --with-pmix.

However, as I stated, I'm using the slurm-wlm package which seems to not
include the pmix functionality by default.

The other option provided:
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.

Similar issue, not building slurm from source doesn't include the PMI
library. I've installed some develop level packages, including the
libpmi2-0 package <https://packages.debian.org/buster/libpmi2-0> which
didn't seem to actually install anything useful as far as I can tell using
the "find" command.

It's sort of looking like I should be looking at building slurm from source
again, I guess.

Thanks,

~Avery Grieve
They/Them/Theirs please!
University of Michigan


On Thu, Dec 10, 2020 at 11:16 AM Christopher J Cawley <ccawley2 at gmu.edu>
wrote:

> I have a 7 node jetson nano cluster running at home.
>
> Send me what you want me to take a look at .  If it's not
> a big deal, then I can let you know.
>
> Ubuntu 18 / slurm <some version from rpm>
>
> Thanks
> Chris
>
>
> *Christopher J. Cawley*
>
> *Systems Engineer/Linux Engineer, Information Technology Services*
>
> *223 Aquia Building, Ffx,** MSN**: 1B5*
>
> *George Mason University*
>
> *Phone:** (703) 993-6397*
>
> *Email:* *ccawley2 at gmu.edu*
> ​
>
> ------------------------------
> *From:* slurm-users <slurm-users-bounces at lists.schedmd.com> on behalf of
> Avery Grieve <agrieve at umich.edu>
> *Sent:* Thursday, December 10, 2020 10:51 AM
> *To:* slurm-users at lists.schedmd.com <slurm-users at lists.schedmd.com>
> *Subject:* [slurm-users] slurm-wlm package OpenMPI PMIx implementation
>
> Hi Forum,
>
> I've been putting together an ARM cluster for fun/learning and I've been a
> bit lost about how to get OpenMPI and slurm to behave together.
>
> I have installed the slurm-wlm package
> <https://secure-web.cisco.com/14EwNb3UZYABzVqRN7IxszUw4L04o_2Bv7wm3a5vivtuqZhDuY3UrhulGE47J31qdoC16rhtMefWeyLXhK10TMim7oOCehTuBJR_47pTBDKcO_xYDX3yqOG1yzamsO31hXo3HS9tSUpOssM40vTLwy4Mxfggu2Qu_yXjJqtLE43mV2CrECvinY7hMt_cRMzi4b8xrKZXqngR31DMmyA9DzimeyLsN7nwxh6kJRMhcg2MjHlCOhu356VVZrErEM9ZafOD66sDUMluigARg1icclZaJOLhXE-7PlFRtAdk2dhXLEvRqSL3SUKrVeBy01MCmSi7sH8bkIijrujncTBU-DfWxY_JOwqhhsJAyXl0XJgjoOiGWHKcLPRRvrCbn_SGHGSw2Ogq3aC4sJLY1tBLwpgvXcOxFoURgb6y6WfJJg04H9ewyQ-Azr7kA_en7DIk_4KOux310uOWzo7XrHTxnLg/https%3A%2F%2Fpackages.debian.org%2Fbuster%2Fslurm-wlm>from
> the Debian apt search and compiled OpenMPI from source on my compute nodes.
> OpenMPI has been compiled with the option --with-slurm and the configure
> time log indicates openmpi has pmix v3 built in. I thought that would be
> enough for slurm and calling a job with "srun -n 4 -N1 executable" (with
> slurm.conf having MpiDefault=pmix_v3) would be enough.
>
> Not the case, unfortunately as slurm doesn't have any idea what pmix_v3
> means without being compiled against it I guess. I have also attempted to
> compile openmpi from source with the --with-pmi option but the slurm-wlm
> package doesn't install any of the libraries/headers (pmi.h pmi2.h pmix.h
> etc). Neither does any of the slurm-llnl develop packages, so I'm at a loss
> of what to do here.
>
> A few notes: OpenMPI is working across my compute nodes. I'm able to ssh
> to my compute node and start a job manually with mpirun that executes
> successfully across the nodes. My slurmctld and slurmd daemons work for
> single thread resource allocation (and presumably OpenMP multithreading,
> though I haven't tested this).
>
> Beyond compiling slurm from source (assuming this installs the pmi headers
> that I can use to build openmpi), which I have tried with no luck on my
> devices, is there a way to get slurm and openmpi to behave together using
> the precompiled package slurm-wlm?
>
> Thank you,
>
> ~Avery Grieve
> They/Them/Theirs please!
> University of Michigan
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20201210/a00f843a/attachment.htm>