0

I am trying to run WRF (real.exe, wrf.exe) through the crontab using compute nodes but compute nodes are not able to run slurm job. I think there is some issue with the MPI library when it's running through the cron environment.

Tried to replicate the terminal path variables while running the crontab job. The log file generated while running job on terminal and crontab is attached as with_terminal and with_crontab respectively in the link. https://drive.google.com/drive/folders/1YE9OchSB8alpZSdRl-8uIbBPm6lI-0DQ

Error while running the job from crontab is as follows

#################################
compute-dy-c5n18xlarge-1
#################################
Processes 4
[mpiexec@compute-dy-c5n18xlarge-1] check_exit_codes (../../../../../src/pm/i_hydra/libhydra/demux/hydra_demux_poll.c:117): unable to run bstrap_proxy on compute-dy-c5n18xlarge-1 (pid 23070, exit code 65280)
[mpiexec@compute-dy-c5n18xlarge-1] poll_for_event (../../../../../src/pm/i_hydra/libhydra/demux/hydra_demux_poll.c:159): check exit codes error
[mpiexec@compute-dy-c5n18xlarge-1] HYD_dmx_poll_wait_for_proxy_event (../../../../../src/pm/i_hydra/libhydra/demux/hydra_demux_poll.c:212): poll for event error
[mpiexec@compute-dy-c5n18xlarge-1] HYD_bstrap_setup (../../../../../src/pm/i_hydra/libhydra/bstrap/src/intel/i_hydra_bstrap.c:772): error waiting for event
[mpiexec@compute-dy-c5n18xlarge-1] main (../../../../../src/pm/i_hydra/mpiexec/mpiexec.c:1938): error setting up the boostrap proxies

Thanks for looking in the issue.

Improve this question
4
  • questions should be self-contained. for security reasons, many SO users (including myself) will not click any link to 3rd party websites (such as google drive) Commented Oct 26, 2021 at 6:54
  • Will you please guide me about how can I attach the log files to this question Commented Oct 26, 2021 at 8:36
  • simply edit your question and copy/paste the logs in a code block. Commented Oct 26, 2021 at 9:30
  • Please check here stackoverflow.com/questions/69790257/… Commented Nov 5, 2021 at 5:31
Related questions
0
error on running mpi job
1
MPI program only executed once in slurm script
1
mpirun with slurm : how to run multiple processes on a single CPU
Load 7 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.