Where is the code in the codebase that handles if a job has outlived its ttl?

[iggeehu]

I have been poking around the codebase for a bit, and I think around the vicinity of the Queue.dequeue_any and Job.fetch(), and Job.restore() might have what I am seeking. But I am stuck there.

Any help is appreciated!

In my mind, I am looking for a piece of code to check the job's ttl when it is plucked from queue. Or maybe Job.perform_job() is run and it handles expired jobs internally, but I didn't find anything in perform_job().