Lock isn't deleted if task killed by hard-time limit

[Strawl]

If i set a hard time limit and the task exceeds that limit the task gets killed. But Celery Singleton doesn't delete it, so it keeps it from creating new tasks.

Proof: Hard time limit exceeded:

[2021-10-01 07:33:44,026: ERROR/MainProcess] Hard time limit (300s) exceeded for tasks.<some_task>[04c8d60c-c7c7-43f0-ad8c-0d478da732a0] [2021-10-01 07:33:44,138: ERROR/MainProcess] Process 'ForkPoolWorker-3' pid:3944596 exited with 'signal 9 (SIGKILL)'

Then celery says this task still exists:

Traceback (most recent call last): File "/home/build/dev/ansible-integration/project_overview/env/lib/python3.6/site-packages/celery/app/trace.py", line 515, in trace_task priority=task_priority File "/home/build/dev/ansible-integration/project_overview/env/lib/python3.6/site-packages/celery/canvas.py", line 219, in apply_async return _apply(args, kwargs, **options) File "/home/build/dev/ansible-integration/project_overview/env/lib/python3.6/site-packages/celery_singleton/singleton.py", line 116, in apply_async return self.on_duplicate(existing_task_id) File "/home/build/dev/ansible-integration/project_overview/env/lib/python3.6/site-packages/celery_singleton/singleton.py", line 141, in on_duplicate task_id=existing_task_id, celery_singleton.exceptions.CelerySingletonException: Attempted to queue a duplicate of task ID 04c8d60c-c7c7-43f0-ad8c-0d478da732a0

For now ill try using soft time limits, which should solve my problem all together, that way my tasks can finish up normally (so it is deleted from that celery-singleton storage or whatever). I don't think this is really that big of an issue, just please address this somewhere so people know.

[steinitzu]

I see, since hard time limit force terminates and doesn't allow for any cleanup.

I think you could set lock_expiry to match the time limit https://github.com/steinitzu/celery-singleton#lock_expiry
The key gets expiry time in redis so it's always deleted

Maybe celery-singleton could also detect the presence of hard timeout and do this automatically.