How do you manually move a job to the failed state and not let RQ retry?

[yawhide]

Hello,

I am trying to use the on_failure callback to prevent a job from retrying. Is there a better way than setting the job's retry_count to 0?
I cant find any documentation or an method I can call to "prevent a job from retrying".

My use case I am trying to solve is to prevent a job from retrying if it fails due to a JobTimeout exception.

thanks

[selwin]

My use case I am trying to solve is to prevent a job from retrying if it fails due to a JobTimeout exception.

This is a good point, we should probably implement something like Retry(max=3, unless=[JobTimeout, MyOtherException]).

[robhudson]

I looked into this as well. One option would be to subclass the Worker or SimpleWorker class and override the handle_job_failure method, adding your logic there.

Example:

class MyWorker(Worker):
    def handle_job_failure(self, job: 'Job', queue: 'Queue', started_job_registry=None, exc_string=''):
        if <logic-to-abort-job-retries>:
            self._stopped_job_id = job.id
            job.retries_left = None
        super().handle_job_failure(job, queue, started_job_registry, exc_string)

[selwin]

Besides implementing customizable exceptions like Retry(max=3, unless=[JobTimeout, MyOtherException]), we can also introduce an RQ specific FailedJobError (better name suggestions welcome) which when raised will be moved to FailedJobRegistry without being retried.

I'm happy to accept a PR for both :)