Umbrella issue.
Most of the mungers can be migrated to Prow plugins and respond to github webhook events but a few need to run periodically without having received an event. These should probably be implemented as kubernetes cron jobs.
EDIT: here's the munger list as of 2018-08-27
| munger | plan | status|
| --- | --- | --- |
| ~approval-handler~ | prow/plugins/approve | done (#5115) |
| assign-fixes | no replacement [3] | n/a |
| block-path | prow/plugins/blockade | done (#4944) |
| blunderbuss | prow/plugins/blunderbuss | done (#5547) |
| check-labels | label_sync | done |
| ~bulk-lgtm~ | remove | done (#4101) |
| cherrypick-auto-approve | [1] | |
| cherrypick-clear-after-merge | [1] | |
| cherrypick-must-have-milestone | [1] | |
| cherrypick-queue | [1] | |
| ~close-stale~ | prow/plugins/lifecycle + cron job | done (#5707) |
| comment-deleter | no replacement [2] | n/a |
| docs-need-no-retest | prow/plugins/docs-no-retest | done (#4895) |
| ~flaky-job-reporter~ | robots/issue-creator | done (#3970, #4450) |
| inactive-review-handler | no replacement [3] | n/a |
| ~issue-creator~ | robots/issue-creator | done (#3970, #4450) |
| issue-triager | no replacement [3] | n/a |
| label-unapproved-picks | prow plugin | done (https://github.com/kubernetes/test-infra/pull/9128) |
| lgtm-after-commit | external prow plugin | done (https://github.com/kubernetes/test-infra/pull/6886) |
| milestone-maintainer | prow plugin | in-progress (https://github.com/kubernetes/test-infra/pull/9158) |
| needs-rebase | external plugin | done (#6022) |
| nag-flake-issues | no replacement [3] | n/a |
| old-test-getter | no replacement [3] | n/a |
| owner-label | prow plugin | done (#6033) |
| path-label | prow plugin | done (#6033) |
| ~publisher~ | kubernetes/publishing-bot | done |
| ~release-note-label~ | prow/plugins/releasenotes | done (#4718) |
| sig-mention-handler | prow/plugins/requiresig | done (#5662) |
| ~size~ | prow plugin | done (#3794, #4157) |
| stale-green-ci | no replacement [4] | n/a |
| stale-pending-ci | no replacement [3] | n/a |
| submit-queue | prow/cmd/tide | in-progress (#3866) |
| ~triage-filer~ | robots/issue-creator | done (#3970, #4450) |
[1] - The cherrypick process on github should be rethought and simplified if possible.
[2] - The comment-deleter is no longer needed if plugins that create comments remove stale comments at that time.
[3] - The assign-fixes, inactive-review-handler, issue-triager, old-test-getter, nag-flake-issues, and stale-pending-ci are no longer in use so shouldn't be migrated unless they are needed.
[4] - The stale-green-ci munger could be reimplemented as a commenter job (eg: periodic-test-infra-retester) but we don't plan on adding this back unless it is sorely missed
Is there an issue for migrating approvals into a prow plugin? Has anybody started working on that? I am interested in pursuing it next week.
I don't think there is an issue specific to that munger yet, just this issue. No one is working on migrating the approval-handler yet, so that would be great!
discussed during sig-testing today; some mungers can be moved out, some can't (or we'd like to obviate, eg: submit-queue for tide)
will update this issue with a full list
We've got a migration of docs-only-retest-not-required incoming
@spiffxp I think we can mark off the release note munger (?) :D
@BenTheElder Just deployed the changes :smile: I'll mark it.
https://github.com/kubernetes/test-infra/pull/5010 may end up being a suitable replacement for the existing cherry-pick mungers
Alternatively, it could just continue to create the cherry-pick PR and allow for the current mungers (or their Prow replacements) to layer on top as they had in the past.
The cherry-pick bot doesn't really replace the cherry-pick mungers because the mungers don't handle the creation of cherry-pick PRs.
However, the cherry-pick bot does replace the hack/cherry_pick_pull.sh script and will be much easier to use.
/assign @marun
per sig-testing v1.9 roadmap
Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
If this issue is safe to close now please do so with /close.
Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale
Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
If this issue is safe to close now please do so with /close.
Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale
/remove-lifecycle stale
The above table has gone stale, I'll update it. The submit queue still exists, so this isn't done yet
/milestone 1.12
@spiffxp The table needs an update for label-unapproved-picks. :)
@nikhita done, and thanks again
/close
We have removed mungegithub https://github.com/kubernetes/test-infra/pull/9362
@spiffxp: Closing this issue.
In response to this:
/close
We have removed mungegithub https://github.com/kubernetes/test-infra/pull/9362
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.
Most helpful comment
@BenTheElder Just deployed the changes :smile: I'll mark it.