Closed luator closed 7 months ago
Alternatively (maybe even the better solution) cluster_utils should detect if a job is killed from outside (i.e. via condor_rm
). Currently, this does not seem to be the case.
By Felix Widmaier on 2022-06-03T09:01:04 (imported from GitLab)
I implemented a stop_job
command (it was actually pretty simple). Needs a bit of testing but I will open a PR for it soon.
By Felix Widmaier on 2022-06-08T08:53:44 (imported from GitLab)
Duplicate of #17
In the interactive mode, there is already a command
stop_remaining_jobs
which cancels all jobs that are currently running. However, I'm often in the situation that there are one or two jobs which are somehow hanging, so I would like to cancel them without impairing the other running jobs. Therefore I think a great addition would be astop_job
command, where one needs to specify a job ID and then only the specified job is aborted.I haven't looked into it yet but would hope that that given the commands that are already there, this might be relatively easy to add.