Draft: revert #849 stale publication handoff#850
Closed
brokemac79 wants to merge 1 commit into
Closed
Conversation
…)" This reverts commit a0143cc.
clawsweeper
Bot
added
rating: 🧂 unranked krab
Contributor
|
Codex review: needs real behavior proof before merge. Reviewed July 25, 2026, 3:08 AM ET / 07:08 UTC. ClawSweeper reviewWhat this changesThis draft changes the exact-review publication workflow and its contract test so a superseded queue acknowledgement fails and retries rather than completing as a terminal handoff. Merge readiness⛔ Blocked until real behavior proof from a real setup is added - 7 items remain Close this PR as superseded by #849. Priority: P1 Review scores
Verification
How this fits togetherThe sweep workflow sends completed exact-review artifacts to the durable publication queue, which either accepts, deduplicates, or supersedes an older publication. That acknowledgement controls whether the workflow finishes or retries, and ultimately whether the review verdict is delivered once by the current publisher. flowchart LR
A[Completed exact review artifact] --> B[Sweep publication step]
B --> C[Durable publication queue]
C --> D{Queue acknowledgement}
D -->|Queued or ordinary dedupe| E[Complete workflow]
D -->|Superseded by newer publication| F[Newer publisher owns delivery]
B -->|Rollback behavior| G[Retry stale publication]
Decision needed
Why: The draft intentionally restores an earlier failure mode as a contingency, but the provided discussion contains no post-merge production evidence that the newer durable publication failed to deliver; choosing retry semantics is an operational policy decision. Before merge
Findings
Agent review detailsSecurityNone. Review metrics
Root-cause clusterRelationship: Members:
Proposal only: this assessment does not dispatch repair, suppress jobs, mutate sibling items, close, or merge anything. Merge-risk optionsMaintainer options:
Technical reviewBest possible solution: Close this PR as superseded by #849. Do we have a high-confidence way to reproduce the issue? No high-confidence reproduction of a bad current-main outcome is provided. The source clearly shows that this branch restores failure-and-retry handling for superseded acknowledgements, but no production evidence shows that the newer publisher failed to deliver. Is this the best way to solve the issue? No. A direct revert restores the retry behavior that #849 addressed; the safer path is to preserve terminal handoff and investigate any verified delivery gap separately. Full review comments:
Overall correctness: patch is incorrect AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against a0143ccfd96d. LabelsLabel changes:
Label justifications:
EvidenceWhat I checked:
Likely related people:
Rating scale
Overall follows the weaker of proof and patch quality. Workflow
|
Contributor
|
Thanks @brokemac79 for preparing this contingency. This rollback would turn an explicit superseded-publication acknowledgement back into a failed workflow result, restoring the retry loop while the newer durable publication already owns delivery. There is no incident showing the merged handoff behavior is broken, so we are closing this draft. Please open a fresh revert PR if an actual production incident demonstrates that a rollback is needed. |
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.
Draft rollback for #849
This is a draft-only rollback of #849, requested as a prepared fallback after #849 merged as
a0143ccfd96don 2026-07-25.What it reverts
Restores the #836 handling in which an
ok: true,deduped: true,superseded: trueexact-review publication acknowledgement is visible as a failed workflow outcome rather than a successful handoff.Current status — do not merge
Codex review found a P1 against this rollback: it can restore retry churn for a stale publication while the newer durable publication owns delivery. This branch is intentionally a draft fallback only; it requires maintainer review and evidence of a bad #849 production outcome before any merge decision.
Validation
pnpm run build:allpnpm exec node --test --test-name-pattern "exact review publication enqueue rejects a superseded acknowledgement" test/sweep-workflow.test.tsgit diff --check origin/mainThe reverted #836 formatting is not accepted by the current formatter check; this draft preserves the exact revert without unrelated reformatting.
References