Repository navigation
🐛 fix: preserve supersession decisions across stage reconciliation - #14979
Conversation
Keep a supersession cause label separate from CI and author repair state, retain escalation evidence in tracker audit replay, and expose the closer and keep-open reason even when a stale record still says 071/072. Allow trusted maintainer decisions to close superseded contributor PRs without a branch push. Detect held duplicates without writes, allow ancillary file differences only with implementation overlap, and skip closed contributor queue snapshots. Hive-Run: #14814 Hive-Plan: issue-14814-supersession-sweep-and-stage-tracker Hive-Spec: issue-14814#fleet-generic-fix Co-authored-by: clubanderson <407614+clubanderson@users.noreply.github.com> Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
Persist supersession park labels in the dashboard sweep fixture and assert the second pass makes no additional writes. Preserve slash-containing label names in the GitHub fixture so reopened-issue cleanup is observed. Signed-off-by: thatguywob <thatguywob@users.noreply.github.com>
…e reject routable TestReviewQueue_SweepDetectionWithoutAutoClose built its client without a stage fact sink, so the new escalation path re-read the PR through the live reconciler and hit the fixture's 404 (dashboard shards 4/4 and shuffle 3/3). Install the recording sink the pkg/github sweep tests use and assert the Parked fact. A bare `/hive decision reject` on a PR nothing supersedes now continues to the changes-requested path instead of being answered with "stays open"; only the explicit `close superseded by #N` form is always answered by the supersession handler. Covered by a sweep-level regression test and the maintainer-commands doc. Hive-Fix-Attempt: 1/6 Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Signed-off-by: ci-maintainer <280983584+hivecommons-hive[bot]@users.noreply.github.com>
kubestellar-prow
Bot
added
the
dco-signoff: yes
kubestellar-prow
Bot
added
the
size/XL
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
hivecommons-hive
Bot
added
author/hive
![hivecommons-hive[bot]](https://avatars.githubusercontent.com/in/3568013?s=60&v=4){kind=small}
There was a problem hiding this comment.
Decision needed: @clubanderson should decide whether #14821 (contributor fork branch, @thatguywob) or this hive adoption #14979 carries commit beb4233 for #14814 — both heads are the identical SHA, so the other should be closed.
Decision needed: @clubanderson should decide whether #14821 (contributor fork branch) or this hive adoption #14979 carries commit beb4233 for #14814 — the two heads are the same SHA and the other should be closed.
Facts I verified, not inferred:
hive-gh pr viewshowsheadRefOidbeb4233d8f4beddf313aa72671d4957368e6a58aon both #14821 and #14979, same basev5, same ten-file set (src/pkg/dashboard/stage_tracker.go,api_review_queue_superseded.go,automerge_stage_labels.go, …). This is not a "same files, different fix" pair — it is literally one commit under two PR numbers.- The open decision question on #14821 asks you to choose between it and adoption #14862 (head
c47eff3, "tests red"). #14862 is now closed-unmerged, and this adoption re-uses #14821's head unchanged, so the "different fixes for bare/hive decision reject" conflict that question described no longer exists. That question can be considered answered by events; the only live question is the one above. - The adoption body says CI "went red" on #14821. At this head every check on both PRs is
SUCCESSexceptbuild (linux/arm64…)still in progress andtidepending — identical on both. So neither PR is currently preferable on CI grounds.
Trade-off for the decision: #14821 keeps @thatguywob as PR author (DCO-signed commits are preserved either way); #14979 is a branch the hive fix agent can push to if a rebase is needed. I have not re-reviewed the code at this head — the review already recorded on #14821 at beb4233 applies to this PR verbatim. I have not closed anything.
— hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
There was a problem hiding this comment.
Decision needed: @clubanderson should decide whether to close contributor PR #14821 in favour of this hive adoption #14979 — both sit at the identical head beb4233 and cannot both merge — now that the earlier conflicting adoption #14862 has been closed.
Decision needed: @clubanderson should decide whether to close contributor PR #14821 in favour of this hive adoption #14979 — both sit at the identical head beb4233 and cannot both merge — now that the earlier conflicting adoption #14862 has been closed.
This PR and #14821 are the same bytes. headRefOid is beb4233d8f4beddf313aa72671d4957368e6a58a on both; this branch hive/adopt-14821 is, per its own body, #14821's head "unchanged". So this is not a conflict and not a divergence — it is one change under two numbers.
What changed since #14821 was parked at stage/085-needs-decision:
- The question parked there was "this PR vs its adoption #14862 (head
c47eff3), which carry different fixes for bare/hive decision reject". #14862 is now closed, unmerged (closed_at 2026-10-28T12:40:02Z), so that question is moot. - The adoption body says this was adopted "after CI went red". At this head, checks are 27/29 green with only
build (linux/arm64, …)still in progress andtidepending — nothing red. I note it as a fact, not a complaint; it just means there is nothing for the fix agent to fix yet.
On the code itself I have nothing to add: head beb4233 was reviewed on #14821 by the adjudicator at this exact SHA with no correctness findings, and the only finding then was the #14862 divergence described above. Re-posting that review on the same bytes would be noise.
Recommendation, for a human to act on: keep one. This copy lives on a hive-owned branch (hive/adopt-14821) the fix agent can push to; #14821 is a fork branch the hive can only comment on. If you prefer the contributor's PR to carry the credit, the reverse works equally well — the diffs are identical. I have not closed anything.
— hive: agent=reviewer-2 backend=copilot model=claude-fable-5.1 copilot=1.0.88
hivecommons-hive
Bot
added
stage/085-needs-decision
|
Needs a decision, not a merge:
Head: verdict requires_human @beb4233, 67/68 checks green, 1 pending, DCO yes, soak 0s of 30m0s. What to do: approve this head ( How to release it:
Clear the cause and the hive re-evaluates on its next pass; the stage moves to |
|
Supersession sweep found that this PR's claimed issue is already closed by another merged PR.
This PR touches files the merged PR did not, so the hive is resolving it itself: updating head beb4233 onto This comment is edited in place by the supersession sweep; it is not duplicated. |
hivecommons-hive
Bot
added
stage/050-review-pending
There was a problem hiding this comment.
Looks correct to me, and this is the PR that should carry #14814 — #14821 can be closed in its favour once a maintainer agrees. I have not closed anything.
What this is. Head 3033244 is #14821's head beb4233 (commits 0d69dd4, 30a71c9, beb4233, author sign-offs intact) plus one merge of v5. I diffed both PRs against v5: the two diffs are identical apart from hunk offsets (cmd/hive/main.go, stage_tracker.go, automerge_stage_labels.go moved on v5). Same change, same intent; both cannot merge.
Why this one survives. The decision parked on #14821 was "this PR vs its adoption #14862, which carry different bare-reject fixes". #14862 is now closed, so that question is moot: this adoption carries #14821's own fix, not #14862's. It is on a hive-controlled branch, already rebased, and every check is green on this head (test (dashboard 1/4) and test (dashboard shuffle 3/3), the two that were red at 30a71c9, pass here). #14821 is a fork branch the hive cannot push to and is still sitting at stage/085-needs-decision on the stale question.
Correctness re-read at 3033244. The earlier low-severity finding on 30a71c9 — a non-superseded bare /hive decision reject on a hive-authored PR being swallowed — is fixed: src/pkg/github/pr_supersession_decision.go:40-56 returns handled=false when no live supersession is detected (or the PR is held) and the request was a bare reject, and src/pkg/github/pr_unpark_decision.go:70-80 only returns early when handled is true, so the request falls through to the existing selfAuthored changes-requested path. An explicit close superseded by #N still always gets a reply, including the closer-mismatch case (pr_supersession_decision.go:57-59). The rest of the design (085 cause label surviving CI/rebase/DCO reconciliation, trusted-merger gate at pr_unpark_decision.go:50-52, no contributor-branch push) is unchanged from the version already read on #14821.
— hive: agent=reviewer backend=copilot model=claude-fable-5.1
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
hivecommons-hive
Bot
removed
role/trusted
Supersedes #14821 (adopted after CI went red and the contributor was unavailable; original commits by @thatguywob preserved).
Fixes #14814
This branch is #14821's head
beb4233d8f4bas pushed by @thatguywob (forkthatguywob/hive), unchanged. Last lane reason on the original:contributor-lane:decision-requested.Hive fix agent: rebase this branch on
v5, fix the failing checks / review asks, and push to THIS branch. Keep the original author's commits and theirSigned-off-by; add your own sign-off on your commits. Do not open another PR.