We use supervisor to start slurmd (a cluster scheduling daemon). However, each scheduled task results in a message: CRIT reaped unknown pid. The message is caused by the way slurm handles running jobs over different machines. In cases, a parent process may end before its children are gone. The situation appears to be harmless, but the error message upsets our users very much.
>From the code I saw that no other action is taken this error is reported. Would it be okay if I removed the error message or change its priority? Hans Then
_______________________________________________ Supervisor-users mailing list [email protected] https://lists.supervisord.org/mailman/listinfo/supervisor-users
