Skip to content
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.

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

Workaround Renovate PR force-push spam #161

Closed
wants to merge 1 commit into from

Conversation

cevich
Copy link
Member

@cevich cevich commented Nov 21, 2023

N/B: These changes will affect all containers-org projects setup with Renovate.

Ref:
containers/buildah#5007 (comment)

Attempt to limit excessive Renovate force-push activity related to golang update PRs. This is unlikely a perfect fix, and it may not address the problem at all. I believe there's a renovate and/or configuration bug here, but lack the resources/time to investigate more deeply - this likely requires a temp. reproduction repo. + mock golang package.

Ref:
containers/buildah#5007 (comment)

Attempt to limit excessive Renovate force-push activity related to
golang update PRs.  This is unlikely a perfect fix, and it may not
address the problem at all.  I believe there's a renovate and/or
configuration bug here, but lack the resources/time to investigate more
deeply - this likely requires a temp. reproduction repo. + mock golang
package.

Signed-off-by: Chris Evich <[email protected]>
@cevich cevich requested review from rhatdan and Luap99 November 21, 2023 15:52
@cevich
Copy link
Member Author

cevich commented Nov 21, 2023

@rhatdan @Luap99 PTAL. I'm not super-confident about this, but it's the best I can do at the moment. If either of you have time to explore the issue deeper, that effort would be welcome and appreciated.

I'm calling "force-push spam", when Renovate ignores existing non-renovate commits on an update PR, ignores the stop-updating label, and force-pushes to it's PR sometimes 20+ times in a single day.

Note: We have the global setting rebaseWhen disabled, so the "force-push spam" (at least in the buildah case) is unlikely related to main movement.

Copy link

github-actions bot commented Nov 21, 2023

Successfully triggered github-actions/success task to indicate successful run of cirrus-ci_retrospective integration and unit testing from this PR's 6dfadc42d5b0057e0d8327f8785de1aba46cb313.

@cevich
Copy link
Member Author

cevich commented Nov 21, 2023

Note: There may be existing discussions on this topic upstream. I haven't had a chance to look/search.

@Luap99
Copy link
Member

Luap99 commented Nov 21, 2023

I have not looked if the changes itself look right in the yaml but I like the idea.
Currently there are a lot of rebases on these PRs, it is especially annoying if there are like 3 update PRs then I merge the first one and the other two get rebased even if there was no conflict. While rebasing is a good idea in general I agree that it is to much at the moment.

Now one thing I do not get is where is the timing coming from, i.e. what timezone is this based on? I go over my GH stuff first thing in the morning and start merging these renovate PRs if they are green and I do this before 11 am my timezone and thus before UTC and east cost time so I don't think this will help me personally.

Ideally there would be a option to say lets rebase once a day max or when there is a conflict so it would not depend on the time.

@cevich
Copy link
Member Author

cevich commented Nov 21, 2023

@Luap99 clarification re: rebasing. It's suppose to be disabled (i.e. no rebasing by default) in the shared config. Then individual repos. can enable it if needed (IIRC c/images and c/skopeo turn it on).

What I've seen cause force-push spam, is actual upstream package updates (even very minor ones). The only way I know to limit that is by hour + by schedule 😕

The timezone is configured to UTC. For the schedule strings, I pulled the times from my posterior. I didn't try to pick anything specific, nor try to guess when would be good. If you think there's a better schedule that would work for the team, please let me know.

@cevich
Copy link
Member Author

cevich commented Dec 1, 2023

The update spam seems to have mostly abated w/o this fix.

@cevich cevich closed this Dec 1, 2023
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.

2 participants