[slurm-users] using strigger in case sbatch job submission fails

Mukul Agrawal mukulagrawal78 at yahoo.com
Sat Aug 1 00:09:34 UTC 2020


I manage a cluster in which users submit jobs using sbatch. I consider job submission to be successful when ctldEpilog is executed. I take some administrative actions through ctldEpilog.

Some times there is some problem with sbatch and ctldEpilog is execution not triggered. sbtach exit status still shows successful (0), so that is not very useful. Is there a way to generate a trigger so some administrative action can be taken when this happens? 

I was looking at strigger but I am not sure if that is useful for this purpose. Basically, I want to raise a flag if ctldEpilog does not get executed after sbatch submission for any reason.

Any pointer or help will be appreciated.


More information about the slurm-users mailing list