[8.x] Add a queue:prune-failed command

[themsaid]

This command removes stale records from the failed_jobs database table. You can pass an --hours option with the number of hours to retain jobs data

[netpok]

Shouldn't be better to show an error message and exit with a non-zero return code if the provider does not support pruning instead of silently failing?

[wouterrutgers]

May I ask what the difference is with the queue:flush command? Except for the limit parameter, which is only available in 9.X.

But that would mean, as far as I can tell, that from 9.X onwards there are 2 commands which will do identical things. What am I missing 😜?

[themsaid]

@wouter2203 flush removes everything while prune keeps the latest records.

[wouterrutgers]

@themsaid Ah, my bad, didn't notice the default of 24 hours to keep!

[bryceandy]

2 commands which will do identical things

There was a PR before this where flush uses an age option:

php artisan queue:flush --age=30
// Failed jobs older than 30 days

php artisan queue:prune-failed --hours=30
// Failed jobs older than 30 hours