Skip to content

Fix lost runs in CronTriggerTimetable #32921

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
BohdanSemonov wants to merge 1 commit into from
Closed

Fix lost runs in CronTriggerTimetable #32921

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion airflow/timetables/trigger.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -89,7 +89,7 @@ def next_dagrun_info(
else:
next_start_time = self._align_to_next(restriction.earliest)
else:
start_time_candidates = [self._align_to_next(DateTime.utcnow())]
start_time_candidates = [self._align_to_next(DateTime.utcnow().replace(second=0, microsecond=0))]

@hussein-awala hussein-awala Jul 29, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This makes sense 🤔

Could you add a test for this change? (a test which fails without resetting seconds and microsecond and passes after applying your change)

@potiuk potiuk Jul 30, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice one, yeah. That would definitely explain occasionally lost dag runs

@potiuk potiuk Jul 30, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And yeah - having a test for that would be fantastic - if you need help @BohdanSemonov - we can suggest an approach here.

@uranusjr uranusjr Jul 31, 2023 •
edited
Loading

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wouldn’t this still have problems if the scheduler is a tad too slow and fails to schedule the DAG in a minute? I still prefer the _align_to_prev fix over this.

@potiuk potiuk Aug 6, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@uranusjr - could you please elaborate? Is there a fix in progress you are referring to? Or some new code to be written? It wasn't too clear. I think it would be worth to consider fixing this problem for 2.7.0

@uranusjr uranusjr Aug 7, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The linked issue already contains discussion before this fix was proposed.

@potiuk potiuk Aug 7, 2023 •
edited
Loading

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah. There are 3 linked issues and each of them has tens of comments. Maybe that is obvious to you, but would be great to point to it. A link to specific comment would be great, otherwise it forces somoene looking here to read the entire conversation (and few of those) to find it..

@potiuk potiuk Aug 7, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

(Just try to see the other side of people who are participating in multiple discussions and might not remember all of it).

@CcLarryKomenda CcLarryKomenda Aug 7, 2023

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I believe he means the change shown in these 2 comments. Apologies for interjecting, but I am excited for this fix so wanted to help keep things moving.

#27399 (comment)
#27399 (comment)

@potiuk potiuk Aug 13, 2023

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it relies on @uranusjr -> I am unable to assess if it's good enough or not at this stage and how to move forward.

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:
Expand Down