ci: remove nightly-failure issue creation, let tend-ci-fix handle it - #6124
Merged
Conversation
The create-issue-on-nightly-failure job opens a triage issue (via the prql-bot PAT so it cascades) on every nightly tests failure. That fires tend-triage on issues:opened while tend-ci-fix independently fires on the failed tests workflow_run — two workflows racing on the same root cause, which produced duplicate fix PRs (#5959 / #5960). tend-ci-fix already triggers on the failed tests run and, when it can't produce a fix, now opens a deduplicated tracking issue itself, so the nightly-health signal is preserved without the duplicate trigger. Refs max-sixty/tend#627
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Removes the
create-issue-on-nightly-failurejob (and its now-orphaned.github/nightly-failure.mdtemplate) fromtests.yaml.Why
That job opens a triage issue on every nightly
testsfailure using theprql-botPAT — deliberately, so theissues: openedevent cascades totend-triage(see #5770). Buttend-ci-fixalso fires independently on the failedtestsworkflow_run. Two workflows then run a Claude session on the same root cause with no shared concurrency group, and the dedup recheck can't close a sub-propagation-window race. That's what produced the duplicate fix PRs #5959 / #5960 on the Hugo 0.162.0 nightly break.The one thing that job still gave us — a durable, human-visible "nightly is broken" signal in the issues list — is now covered by
tend-ci-fixitself: when it can't produce a fix it opens a deduplicated tracking issue. So removing this job collapses the duplicate-trigger race cleanly without losing the nightly-health signal:testsfails →create-issue-on-nightly-failureopens an issue →tend-triageandtend-ci-fixboth fire.testsfails →tend-ci-fixfires once; if it can't fix, it leaves a tracking issue.This is the change the thread in max-sixty/tend#627 landed on ("disable the issue creation on test fail, and tend should take care of it"); the upstream tracking-issue behavior it depended on has since shipped in the bundled
ci-fixskill.If the job was kept intentionally — e.g. you prefer the richer
nightly-failure.mdissue for human visibility and accept the occasional tend-auto-closed duplicate — feel free to close this; it's a proposal, not a correction.Refs max-sixty/tend#627