Recover Cron worker from missed tick

[Similiukas]

Is there a way to recover cron worker from a missed tick? For a scenario where the machine is put to sleep, if the scheduled time is missed, worker emits Error::SourceError::OutOfRangeError and then never runs after that.

This happens for example on fly.io machines which can be suspended when there is no load. Or just more simply replicate it running a docker and putting the device to sleep to miss the next schedule. Then the worker executes not on time and .on_event() emits an error, however, the worker successfully completes the job. That is, some_fn() runs to finish without problem. However, next job is not scheduled and the worker never executes again.

Code snippet

Monitor::new()
        .register(
            WorkerBuilder::new("cron-job")
                .retry(RetryPolicy::retries(5))
                .catch_panic()
                .enable_tracing()
                .backend(CronStream::new(Schedule::from_str("0 */5 * * * *").unwrap()))
                .build_fn(some_fn),
        )
        .on_event(handle_worker_events)
        .shutdown_timeout(Duration::from_secs(20))
        .run_with_signal(signal::ctrl_c())
        .await

Testing timeline:

2025-02-13T18:09:59.985510Z  INFO task{task_id="01JM07RJ6HYNWAXA7XQWD6YMEZ" attempt=4}: worker started at 2025-02-13 18:09:59 UTC
2025-02-13T18:10:00.755990Z DEBUG task{task_id="01JM07RJ6HYNWAXA7XQWD6YMEZ" attempt=4}: apalis::layers::tracing::on_response: task.done done_in=770ms result=()
------- DEVICE IS PUT TO SLEEP FOR 15 MINUTES  -------
2025-02-13T18:27:50.717436Z  INFO task{task_id="01JM092CSX2DHXJZ57SRD31CJD" attempt=4}: worker started at 2025-02-13 18:27:50 UTC
2025-02-13T18:27:50.744200Z ERROR worker failed with SourceError(OutOfRangeError(())
2025-02-13T18:27:51.804062Z DEBUG task{task_id="01JM081Q5GETD8FTFFSCHQR5JK" attempt=4}: apalis::layers::tracing::on_response: task.done done_in=870ms result=()

[geofmureithi]

This is resolved in #513