[sumaiazaman]

A unique job scheduled with Schedule::job() keeps its unique lock when pushing it to the queue fails. Once the queue is reachable again, the scheduler keeps skipping the job because the lock from the failed run is still held, and with the default uniqueFor of 0 it never expires:

Schedule::job(new SyncInventory)->everyFiveMinutes(); // SyncInventory implements ShouldBeUnique

// 10:00  push fails, Redis is unreachable
// 10:05  skipped, the lock is still held
// 10:10  skipped

Schedule::dispatchUniqueJobToQueue() acquires the lock and then dispatches the job, but nothing releases the lock if the dispatch throws. This is the same issue #62010 fixed for dispatch() and #62045 fixed for queued listeners, in the remaining place that acquires the lock itself.

This releases the lock when dispatching throws and rethrows the exception, the same way as those two.