You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[shared-ci] Sign-off survives the squash failing across hivecommons/hive #19190
Shared CI baseline triage (hive-baseline-check.sh) returned DEFER_TO_INCIDENT for check Sign-off survives the squash: red on at least three open sibling PRs targeting v5 (siblings: #17512, #19063, #19102, #19110, #19143 as sampled), with the base branch result missing. Most siblings are behind v5. Likely related to #19038 (PR-contents merge check). Affected PRs are deferred until this closes or the baseline turns green.
No open PR against v5 currently has this check red.
The workflow only runs on pull_request, so there is never a base-branch result. That is by design, and the baseline-check script reports it as "missing".
The recent failures (e.g. run 44431224820 on feat/pr-gate-grant-visibility) are the checker correctly rejecting a branch commit whose Signed-off-by doesn't name the PR author. That is a per-PR contributor fix (git commit --amend -s), not a defect in the workflow or script.
Any follow-up belongs to the baseline triage (hive-baseline-check.sh) treating a pull_request-only check as having no base result. It is not a change to this repo's CI. Closing is left to a maintainer.
Sign-off survives the squash was red on each sibling for that PR's own reason. On #19110, #19102 and #19063 the failing commit was a fix-up the hive pushed onto a human-authored PR, signed off as hivecommons-hive[bot] (for example FAIL f65da8c bot-signoff on #19110). A squash merge would have landed it under the PR author with a mismatched sign-off, which is exactly what the check is there to stop. The git fetch failed or stalled text in the logs is the step's script being echoed, not a real fetch failure. The check is green on the default branch's recent PRs (#19196, #19198 passed it).
The triage misfire (check has no result on the default branch and most siblings were behind it, but it still returned DEFER_TO_INCIDENT) is fixed by #19198 (#19191).
Activity
hivecommons-hive commentedon Nov 19, 2026
🔓 Claim by @Danathar released (lease revoked).
Claim history
Danathar(contributor on hive hosted-available-oke-11-placeholder-r05x)Danathar(contributor; lease revoked)Danathar commentedon Nov 19, 2026
Checked on v5 (38c936d): no repository change is warranted here.
Sign-off survives the squash= SUCCESS. 🐛 fix(github): recognise the hive's own unpark replies when it posts as an unconfigured login #17512 has three CANCELLED runs from earlier but a SUCCESS on its latest run.pull_request, so there is never a base-branch result. That is by design, and the baseline-check script reports it as "missing".feat/pr-gate-grant-visibility) are the checker correctly rejecting a branch commit whoseSigned-off-bydoesn't name the PR author. That is a per-PR contributor fix (git commit --amend -s), not a defect in the workflow or script.Any follow-up belongs to the baseline triage (
hive-baseline-check.sh) treating a pull_request-only check as having no base result. It is not a change to this repo's CI. Closing is left to a maintainer.-- hive: backend=claude model=claude-sonnet-5-5
🐝 Hive Agent:
contributor| SHA:a408245b5clubanderson commentedon Nov 19, 2026
Closing: this was not a shared CI incident.
Sign-off survives the squashwas red on each sibling for that PR's own reason. On #19110, #19102 and #19063 the failing commit was a fix-up the hive pushed onto a human-authored PR, signed off ashivecommons-hive[bot](for exampleFAIL f65da8c bot-signoffon #19110). A squash merge would have landed it under the PR author with a mismatched sign-off, which is exactly what the check is there to stop. Thegit fetch failed or stalledtext in the logs is the step's script being echoed, not a real fetch failure. The check is green on the default branch's recent PRs (#19196, #19198 passed it).What was done:
native-queue-policyfor that reason, not this one.The triage misfire (check has no result on the default branch and most siblings were behind it, but it still returned DEFER_TO_INCIDENT) is fixed by #19198 (#19191).