Skip to content

Conversation

@pvchupin
Copy link
Contributor

Task time is ~1h10m.
Will see how it impacts the pool.

Task time is ~1h10m.
Will see how it impacts the pool.
@pvchupin pvchupin requested a review from a team as a code owner September 22, 2022 23:18
@bader
Copy link
Contributor

bader commented Sep 23, 2022

Task time is ~1h10m. Will see how it impacts the pool.

I'm looking at existing pre-commit and I see that it takes ~36 min (19 min build + 17 min longest test), which is twice faster than expected Windows job. At the same time we have 2 times less Windows machines than Linux machines.
It sounds like we are going to have delays in pre-commit results.

@pvchupin
Copy link
Contributor Author

Task time is ~1h10m. Will see how it impacts the pool.

I'm looking at existing pre-commit and I see that it takes ~36 min (19 min build + 17 min longest test), which is twice faster than expected Windows job. At the same time we have 2 times less Windows machines than Linux machines. It sounds like we are going to have delays in pre-commit results.

Yes, that's a drawback I'd like to highlight. Pros: getting Windows issues earlier. Cons: pre-commit will take longer.
Opinions?

@bader
Copy link
Contributor

bader commented Sep 28, 2022

I'd like to note that pre-commit/post-commit CI jobs ratio today is 4/1.
As an alternative, I would consider configuring a notification email with post-commit results to PR author + a mailing list.

@pvchupin
Copy link
Contributor Author

As an alternative, I would consider configuring a notification email with post-commit results to PR author + a mailing list.

That's a reasonable alternative. Do you have any good pointers to explore?

Meanwhile just added a commit to reduce the load, skipping check-llvm and check-clang which take ~half of the time. That should make this approach a little bit more bearable...

Copy link
Contributor

@cperkinsintel cperkinsintel left a comment

Choose a reason for hiding this comment

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

LGTM

@bader
Copy link
Contributor

bader commented Oct 2, 2022

Do you have any good pointers to explore?

No, unfortunately, but I didn't looked for it. AFAIK, GitHub sends notifications to the account, which triggered the event only, so probably it can be some bot or third-party action.

@pvchupin
Copy link
Contributor Author

We found out that the main problem of post-commit today is notifications of the failures. Mail on post-commit broken goes to the person who merged the change, rather than the author. We need to fix that first.

@pvchupin pvchupin closed this Oct 25, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants