I have a compiled binary that will run with srun but not with mpirun.
The attempts to run with mpirun all result in failures to initialize.
I have tried this on one node, and on two nodes, with firewall turned
on and with it off.
Am I missing some command line option for mpirun?
OMPI built from t
Add --enable-debug to your OMPI configure cmd line, and then add --mca
plm_base_verbose 10 to your mpirun cmd line. For some reason, the remote daemon
isn’t starting - this will give you some info as to why.
> On Jun 17, 2018, at 9:07 AM, Bennet Fauber wrote:
>
> I have a compiled binary that
I rebuilt with --enable-debug, then ran with
[bennet@cavium-hpc ~]$ salloc -N 1 --ntasks-per-node=24
salloc: Pending job allocation 158
salloc: job 158 queued and waiting for resources
salloc: job 158 has been allocated resources
salloc: Granted job allocation 158
[bennet@cavium-hpc ~]$ srun ./te
What MPI is SLURM set to use/how was that compiled? Out of the box, the SLURM
MPI is set to “none”, or was last I checked, and so isn’t necessarily doing
MPI. Now, I did try this with OpenMPI 2.1.1 and it looked right either way
(OpenMPI built with “--with-pmi"), but for MVAPICH2 this definitely
Ryan,
With srun it's fine. Only with mpirun is there a problem, and that is
both on a single node and on multiple nodes. SLURM was built against
pmix 2.0.2, and I am pretty sure that SLURM's default is pmix. We are
running a recent patch of SLURM, I think. SLURM and OMPI are both
being built u
Well, this is kind of interesting. I can strip the configure line
back and get mpirun to work on one node, but then neither srun nor
mpirun within a SLURM job will run. I can add back configure options
to get to
./configure \
--prefix=${PREFIX} \
--mandir=${PREFIX}/share/man \
--with
