php artisan queue:flush does not flush the queues

Our log is blowing up due to non-stop retries by the queue workers despite having the retries set to 1 only. This is literally caused by 3 different failed job payloads that are being retried millions of times due to a fatal error.

1. Why aren't the queued jobs just failed after the 1st one as configured?
2. Why doesn't php artisan queue:flush actually flush the queues so it will stop retrying all jobs?

Queue is SQS based so I went into SQS and purged there and then restart the queue workers, but I would have expected the flush command and the retry setting to have done their jobs.

via Chebli Mohamed