Repository navigation
Description
Summary
Dropping strict_required_status_checks_policy (see #1864's decision record) removed the ruleset's demand that every PR be up to date with main — but auto-rebase.yml still merges main into every open PR on every push to main, unconditionally. The ruleset no longer requires it; the automation does it anyway. So the treadmill's driver is still live, one layer down.
Evidence, 2026-09-21
strict was set false at ~02:55Z. Since then, two pushes landed on main, and the cascade fired on both:
13:22:07 push to main (feat: implement issue #1874 …)
13:22:11 auto-rebase.yml event=push success
13:23:12 #1794 "Merge branch 'main' into dev-lead/…" author=Don Petry committer=GitHub
13:23:20 #1770 "Merge branch 'main' into dev-lead/…" author=Don Petry committer=GitHub
Cost of that single cascade — pr-auto-review runs in the 30 minutes either side:
| Window | Runs |
|---|---|
| 12:50–13:20 (before) | 25 |
| 13:22–13:52 (after) | 51 |
Accumulated merge-from-main commits on individual PRs remain high: #1742 has 23, #1770 has 22, #1794 has 16 — the great majority carrying no content change at all.
Correction to an earlier reading
A pre/post run-rate comparison appeared to show an ~86% drop after the ruleset change. That comparison was confounded and should not be relied on. The "before" window contained four PR merges landed by hand within ~20 minutes (four pushes → four full cascades); the "after" window contained a single push. The lower rate mostly reflects fewer merges to main, not a structurally quieter fleet. When a push did occur, the cascade behaved exactly as before.
Why strict alone could not have fixed it
strict and auto-rebase were two independent enforcers of the same invariant. Removing the first leaves the second untouched: auto-rebase.yml triggers on push: branches: [main] and updates open PRs without ever consulting whether being behind main actually blocks anything.
Acceptance Criteria
auto-rebaseupdates a PR from its base only when being behind actually blocks that PR — i.e. when the ruleset requires up-to-date branches, or the PR is queued to merge, or a human/label explicitly requests it. Being merelyBEHINDwith no such requirement is not a reason to push.- The condition is derived from repository configuration, not hardcoded. If
strictis re-armed later (Merge protections are unsatisfiable at N>1 open PRs: one merge to main forces a full re-review of every other PR — ~154 forced re-reviews in 48h, zero merges #1864 records it as a one-field revert), auto-rebase resumes its current behaviour with no further code change. - Genuine conflict resolution is unaffected: a
DIRTYPR still gets the existing rebase/conflict path. This issue is only about the no-op base merges. - A deterministic counter distinguishes base merges that were required from those that were skipped as unnecessary, so the effect is measurable rather than assumed — and so Merge protections are unsatisfiable at N>1 open PRs: one merge to main forces a full re-review of every other PR — ~154 forced re-reviews in 48h, zero merges #1864 AC Optimize review: small-PR and incremental fast paths #4's metric has a second input alongside pr-review re-runs the full tier stack for a conflict-free base merge — carry the prior verdict forward when the diff-vs-merge-base is unchanged #1865's.
- Measured over a window containing at least three pushes to
main, no-op base merges on non-queued PRs fall to zero.
Relationship to the rest of #1864
#1865 stops the re-review that follows a no-op base merge. This issue stops the no-op base merge itself. They are complementary and neither subsumes the other: #1865 protects against any source of a content-identical new SHA, while this removes the main producer of them. Landing only #1865 still leaves every PR accumulating empty merge commits and churning CI; landing only this still leaves re-reviews triggered by other pushes.
References
- Merge protections are unsatisfiable at N>1 open PRs: one merge to main forces a full re-review of every other PR — ~154 forced re-reviews in 48h, zero merges #1864 — parent; the treadmill measurement and the merge-queue decision record
- pr-review re-runs the full tier stack for a conflict-free base merge — carry the prior verdict forward when the diff-vs-merge-base is unchanged #1865 — the re-review half
.github/workflows/auto-rebase.yml—on: push: branches: [main], no condition on necessity
Activity
github-actions commentedon Sep 21, 2026
Thanks for the exceptional write-up — the cascade timing data and corrected pre/post comparison make the root cause extremely clear. One question before we dig into the fix: since
auto-rebase.ymlis a thin caller stub and behavioral changes need to land in the reusable workflow, do you want the fix implemented as a neweligibilitymode (e.g.eligibility: blocking-only) passed viawith:— keeping the defaultallfor repos that still want unconditional updates — or should the reusable auto-detect the rulesetstrictsetting without requiring a caller-side change?don-petry commentedon Sep 22, 2026
Dev-Lead Implementation Plan
Issue: #1881 — auto-rebase still merges main into every open PR on every push, though the ruleset no longer requires up-to-date branches
Scope & architectural boundary
The behavioural gate (deciding whether
update-branchfires) lives entirely in thecentral reusable
petry-projects/.github/.github/workflows/auto-rebase-reusable.yml(
lib/eligibility.sh) — this repo holds only the channel-pinned thin caller stub.github/workflows/auto-rebase.yml(@auto-rebase/v2-next). Per the AGENTS.md"Caller-stub input forwarding across channel pins" guardrail, I will not add a
with:forward for a new eligibility mode the pinned channel does not yet declare/implement — that
gate (AC #1 behaviour, AC #2 config-read, AC #5) must land in the central reusable first
(the #738/#468 pattern), then be promoted onto the channel, then forwarded. That sequence
cannot be completed from this repo.
What can and should land here is AC #4 — the deterministic, config-derived counter
that makes the no-op base-merge volume measurable (the second input to #1864 AC #4,
alongside #1865's
reviews_carried_forward). It lives in this repo's existing auto-rebasemeasurement surface,
scripts/auto_rebase_health.sh, mirroring how #1865's re-review counterlanded here.
Implementation Checklist
scripts/auto_rebase_health.sh: add purestrict_from_branch_rules(readsstrict_required_status_checks_policyfrom the branch-rules JSON — derived from repoconfig per AC Go-live improvements for PR review agent #2, not hardcoded)
scripts/auto_rebase_health.sh: add puresummarize_base_merges— over opennon-draft non-Dependabot BEHIND PRs, classify each base merge as required
(strict on, OR queued to merge, OR carries the explicit request label) vs skippable
(no-op churn). DIRTY PRs untouched (AC Add @claude delegation, auto-merge, and rebase handling #3).
scripts/auto_rebase_health.sh:main()best-effort reads branch rules + addslabels,autoMergeRequestto the existing single PR snapshot;render_reportgains a"Base-merge necessity (auto-rebase still merges main into every open PR on every push, though the ruleset no longer requires up-to-date branches — the treadmill's driver outlived the strict fix #1881)" section emitting
base_merges_required/base_merges_skippabledeterministically.Tests to Write
tests/auto_rebase_health.bats:strict_from_branch_rulestrue/false/absent/emptytests/auto_rebase_health.bats:summarize_base_merges— strict-off all-skippable,strict-on all-required, queued→required, labelled→required, draft/dependabot/DIRTY excluded, empty→0/0
tests/auto_rebase_health.bats:render_reportsurfaces the base-merge-necessity section + countersTest Command
bats tests/auto_rebase_health.bats(full:scripts/run-bats.sh tests/*.bats) · lint:bash .dev-lead/scripts/dev-lead-lint.shdon-petry commentedon Sep 22, 2026
Dev-Lead: Progress (in progress — not yet durable)
Plan Execution (intended)
measurement surface (
scripts/auto_rebase_health.sh) — the AC Optimize review: small-PR and incremental fast paths #4 deliverable that canlive in this thin-caller repo, mirroring where pr-review re-runs the full tier stack for a conflict-free base merge — carry the prior verdict forward when the diff-vs-merge-base is unchanged #1865's
reviews_carried_forwardlanded..github/workflows/auto-rebase.yml. Per AGENTS.md "Caller-stub input forwarding acrosschannel pins", the actual gate (AC test issue from agent #1 behaviour / AC feat: add Copilot engine support via REVIEW_ENGINE toggle #5) belongs in the central
auto-rebase-reusable.ymland must be landed there, promoted onto the@auto-rebase/v2-nextchannel, and only then forwarded — the [Phase 1] Restrict auto-rebase fan-out to review-ready PRs (free, plan-independent mitigation) #738/fix(pr-review): count only non-converging review cycles; make escalation recoverable #468 pattern. Thatsequence cannot complete from this repo, so it is noted rather than forced.
Scope note (why this half lands here)
strict_from_branch_rulesreadsstrict_required_status_checks_policyfrom the live branch ruleset — verified readingfalseonmainright now (matches the issue premise). If strict is re-armed, everyBEHIND PR flips to
requiredwith no code change.path is unaffected.
summarize_base_mergesemitsbase_merges_required/base_merges_skippable(always both, even zero) over opennon-draft non-Dependabot BEHIND PRs.
skippable= the no-op churn auto-rebase still merges main into every open PR on every push, though the ruleset no longer requires up-to-date branches — the treadmill's driver outlived the strict fix #1881 targets; it isthe second input to Merge protections are unsatisfiable at N>1 open PRs: one merge to main forces a full re-review of every other PR — ~154 forced re-reviews in 48h, zero merges #1864 AC Optimize review: small-PR and incremental fast paths #4.
reusable change described above.
Local check output (provisional — not the pushed run)
Files touched (this workspace, not yet pushed)
scripts/auto_rebase_health.sh: addstrict_from_branch_rules+summarize_base_mergespure helpers; new "Base-merge necessity (auto-rebase still merges main into every open PR on every push, though the ruleset no longer requires up-to-date branches — the treadmill's driver outlived the strict fix #1881)" report section;
main()reads the branchruleset (best-effort, degrades to strict=false with a
::warning::) and addslabels,autoMergeRequestto the existing single PR snapshot.tests/auto_rebase_health.bats: 16 new unit tests for the two helpers and the report section.Notes
contents: readon
auto-rebase-health.yml; unreadable config degrades gracefully.auto-rebase-reusable.yml, promotethe channel, then forward — after which AC feat: add Copilot engine support via REVIEW_ENGINE toggle #5's skippable count should fall to zero.
don-petry commentedon Sep 22, 2026
Dev-Lead: Implementation Complete — PR #1887
The implementation for issue #1881 is durable: commits are pushed and a pull request is open. This record is backed by verifiable artifacts you can confirm in one click.
66de51cb556ee942713a453797f7358cfb97447amain: 2 file(s), +255/-7 linesReview happens on the PR. Acceptance criteria and test results are evidenced by the PR's diff and CI checks — not asserted here. If a later stage fails, this claim is retracted in place.
feat: implement issue #1881 — auto-rebase still merges main into ever…
don-petry commentedon Oct 9, 2026
Status 2026-10-09: PR 1887 (merged) added only the measurement counter (scripts/auto_rebase_health.sh). The actual gate, which is this issue's AC 1, 2 and 5, belongs in the central auto-rebase-reusable.yml and was not tracked anywhere in the public repo. It is now tracked as combsmith/.github#1272 (generic, since that repo is public). This issue stays open until that gate has landed and been promoted, and the base-merge-only share measured by the counter has fallen. Note PRs here still receive 'Merge branch main' commits today (for example on two open dev-lead PRs this afternoon), so the producer is still live.
don-petry commentedon Oct 9, 2026
Correction 2026-10-09, now that the merge queue is enabled (see #1864): disabling auto-rebase outright is NOT safe. Two dependencies found. (1) scripts/dev-lead-intent.sh treats a comment containing '<!-- auto-rebase-conflict:' as its highest-priority event and starts the automatic conflict-resolution (rebase) run, so auto-rebase's failed update attempt is the only thing that detects a conflicting PR and triggers recovery; disabling it would leave DIRTY PRs unrecovered. (2) scripts/dev-lead-fix-issue.sh pre-applies the auto-rebase:ready label to every PR it opens (to avoid the #711 deadlock), so a gate that treats that label as an explicit request would never skip an agent PR. The corrected design, tracked in combsmith/.github#1272: skip the push for cleanly mergeable PRs unless strict or a queue requires it, but keep detecting conflicts WITHOUT pushing and keep posting the exact same conflict notice once per base SHA; the pre-applied label alone no longer forces an update. This issue stays open until that has landed and been promoted.
don-petry commentedon Oct 11, 2026
The central gate is live here; first measured push
The fix for this landed centrally as "auto-rebase updates every behind PR on every push even where being behind blocks nothing" (
petry-projects/.github#1272, merged 2026-10-10 23:27 UTC as98c8a23b). This repository pinsauto-rebase/v2-next, and the scheduled cut moved that tag to98c8a23bat about 23:47 UTC, so no change was needed in this repository.First push to
mainafter the cut (4748e457, 2026-10-11 00:22 UTC), run 38098230072:mainnone applied (strict up-to-date not required))For comparison, the last push before the cut (23:33 UTC, old tag) pushed a base merge to every PR that was behind.
Against the criteria here:
strictrestores the old behaviour with no code change.updating branch [gate: …]orskipping update [gate: …]), which is what the table above counts.Removing the
dev-leadlabel so no second implementation is started in this repository.don-petry commentedon Oct 11, 2026
Criterion 5 measured: three pushes, zero no-op base merges
Every push to
mainsince the gate reached this repository's channel, counted from the per-PR log lines of eachauto-rebaserun:4748e45763c03b4fa44bb9bb49 updates that the old behaviour would have pushed were skipped; 11 attempts were made, all on PRs GitHub reports as conflicting (the existing conflict path, which cannot create a commit); none was a no-op base merge. No run fell back to "could not evaluate".
All five criteria are met, with the one recorded difference from the wording of criterion 1 (the base branch's rules are the only condition; merge-queue membership, auto-merge and labels are not). Closing.
Scope note: this repository has it because it pins the
nextchannel of auto-rebase. The other consumer repositories get it as that channel promotes through its rings.