[12.x] Prevent Queue Worker crash if cache store is unavailable

[jackbayliss]

My DB and app are managed by different companies.

My DB sometimes has maintenance, at which point I schedule the app to go into maintenance during those hours- but the workers always still try to get the cache as they're restarted by the managed provider, regardless of maintenance mode etc.

This is due to getTimestampOfLastQueueRestart which results in me seeing loads of:

SQLSTATE[HY000] [2002] Connection refused (Connection: mysql, SQL: select * from `cache` where `key` in (app_cache_illuminate:foundation:down))

This PR basically makes the queue worker fail gracefully - and exit when the cache is unavailable, but still report the error.

I've also added the ability to also alter the cache exit code, like $memoryExceededExitCode

I've targeted 12.x as I don't think it's a breaking change, but can change it to target 13.x if i'm missing something..

Feel free to adjust etc, as maybe there's a better solution. 🫡

Thank you!

[github-actions]

Thanks for submitting a PR!

Note that draft PR's are not reviewed. If you would like a review, please mark your pull request as ready for review in the GitHub user interface.

Pull requests that are abandoned in draft may be closed due to inactivity.

[taylorotwell]

I probably won't change this behavior on a patch release.

[jackbayliss]

@taylorotwell Thanks, I'll target 13.x once the other worker bits are eventually merged into main.