On Oct 29, 2014, at 9:51 AM, Ralph Castain <r...@open-mpi.org> wrote:
>> I submit the job through command "salloc -N 10 --no-kill mpirun >> ./my-mpi-application". >> >> Hello, you are not supposed to use mpirun with slurm but directly srun (or >> something similar). > > That simply isn’t true - there is no problem using mpirun with Slurm, and > many people do so because they want the options offered by mpirun. Case in point: I use SLURM to manage my development cluster here at Cisco. I salloc to get nodes all the time, and then use "mpirun" inside that allocation to run Open MPI jobs. I also use sbatch to submit scripts that contain "mpirun" to run Open MPI jobs. If your job is being killed by signal 9, then some external entity (to Open MPI) is killing it -- perhaps SLURM itself, or perhaps some other entity (e.g., the OOM killer). -- Jeff Squyres jsquy...@cisco.com For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/
