When a job is forcibly cancelled by calling stop(), the added schedule to the scheduler will no longer create new job.

[Trung0246]

Things to check first

• I have checked that my issue does not already have a solution in the FAQ

• I have searched the existing issues and didn't find my bug already reported there

• I have checked that my bug is still present in the latest release

Version

4.0.0a4

What happened?

How the schedule was created:

	
TRACK_ENGINE = sqlalchemy.ext.asyncio.create_async_engine(f"sqlite+aiosqlite:///{schedule_path}")
pathlib.Path(schedule_path).parent.mkdir(parents=True, exist_ok=True)
TRACK_DB = apscheduler.datastores.sqlalchemy.SQLAlchemyDataStore(
	engine=TRACK_ENGINE,
)

async with apscheduler.AsyncScheduler(data_store=TRACK_DB) as scheduler:
	TRACK_SCHEDULE = scheduler

await TRACK_SCHEDULE.add_schedule(
	track_handle,
	apscheduler.triggers.interval.IntervalTrigger(
		weeks=data.get("weeks", 0),
		days=data.get("days", 0),
		hours=data.get("hours", 0),
		minutes=data.get("minutes", 0),
		seconds=data.get("seconds", 0),
		microseconds=data.get("microseconds", 0),
		start_time=data.get("start_time", datetime.datetime.now()),
		end_time=data.get("end_time", None),
	),
	max_running_jobs = 1,
	conflict_policy = apscheduler.ConflictPolicy.do_nothing
)

The schedule termination mechanism:

await asyncio.gather(TRACK_SCHEDULE.stop(), TRACK_SCHEDULE.wait_until_stopped())

How can we reproduce the bug?

This only happens when the job is RUNNING, not when it's finished and therefore we can safely cancel during that time span.

Use the above code to plug in appropriate order. When TRACK_SCHEDULE restarted with new python runtime instance, the job no longer getting created.

The problem is probably this:

Looks like running_jobs is still at 1 despite the job being cancelled.

[agronholm]

Can you reproduce this on master? I've made TONS of changes around this area, and the problem may not exist anymore. If it does, would you mind creating a minimal working example for me?