Cannot cancel job with 'CG' status

IFB Core Cluster Demande de logiciel ou de banque de données
1

Hello team,

A couple of days ago, I submitted a job array. One of the jobs therein (19598341_151), seems to fail somehow. It is reporting it has the status 'CG' (Job is in the process of completing. Some processes on some nodes may still be active.), but even if I try to cancel the whole array with scancel 19598341, it wouldn't be suspended. Possibly there's some I/O issue. I was wondering whether you are capable of cancelling it for me.

Thanks in advance!

Best regards,
Jolien van Hooff

2

Hello Jolien,

Yes, 'CG' is for "COMPLETING" (Slurm Workload Manager -).
There is an error but I can't cancel this job too.
We have drain the node and we are waiting to reboot the node (to release your job).

Best regards

3

Hi!

Thanks for your response. I'm sorry to hear that. In case you will find out what went wrong exactly, please let me know, then I'll try to avoid that in the future. From my side, I have not been able to pinpoint the error yet.

Best regards,
Jolien van Hooff