uranusjr commented on issue #27399:
URL: https://github.com/apache/airflow/issues/27399#issuecomment-1533666824
This is the logic to calculate the next run for `catchup=False`, for
reference:
```python
start_time_candidates = [self._align_to_next(DateTime.utcnow())]
if last_automated_data_interval is not None:
start_time_candidates.append(self._get_next(last_automated_data_interval.end))
if restriction.earliest is not None:
start_time_candidates.append(self._align_to_next(restriction.earliest))
next_start_time = max(start_time_candidates)
``
The two `if` blocks should be fine since they match the logic for
`catchup=True`. So I wonder if the problem is from the `max` call—maybe under
some edge cases that `max` would wrongly select the `next(utcnow)` result
(first line) instead of the actual schedule it should use—when the scheduler
took too much time to schedule the previous run, perhaps?
Also worth noting that `CronDataIntervalTimetable` (the old one) uses
slightly different logic to account for the current time, and that timetable
seems to be bug-free.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscr...@airflow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
