Repository navigation
v1.94.26.0 gstack-evidence: sum sharded bun lanes; docs: dispatch-carry warning and record-ci recheck order (GSTA-258 L6 follow-up) - #3934
Merged
Conversation
…CI or the ledger (GST-1186) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…ulate proof, policy-base check on carried (GST-1189, GST-1190) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…tree.json records skipped_files (GST-1190) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
… find --policy-from head --simulate (GST-1190) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…rom #3923 per GSTA-258) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…(gate inputs, passed ignores CI rows, simulate consent, pull_request-only skips, guard not-owed) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…nnot print Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…h per_file Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…ns must still stand (l6c-p2) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…'s count must come from this run's row (l6c-p2 red-team) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…check ratchet Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…iles with per_file" This reverts commit fafbae5.
…r_file move to the adoption PR; Step 17 keeps its refs/heads marker Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…oubleshooting (docs audit l6cdocs2) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…mainder run recheck (#3929 second fix-check) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…s: dispatch-carry warning and record-ci recheck order (GBEV-3, GSTA-258 L6 follow-up) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…l; sum only shards 1..N in order; soften the dispatch-carry mechanism claim (l6f-p1 fixes) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
4 tasks done
…ispatch fix (TRUT-7) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
4 tasks done
…owup Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com> # Conflicts: # docs/pregate.md # lib/pregate/index.ts # lib/pregate/run-check.ts # lib/run-evidence.ts # test/ship-ci-carry.test.ts
|
😎 This pull request was merged. |
3 tasks done
Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
…ase-moved fix (the real tree_mismatch path; #3934 fix-check) Co-authored-by: capy-ai[bot] <230910855+capy-ai[bot]@users.noreply.github.com>
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
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.
Summary
bun test --shard=i/Nprocesses now records the sum of the shards' counts, not the last shard's (GBEV-3, gbrain-evals#317: 3336 recorded against 6,095 run across 2 shards). The sum needs shards 1/N to N/N, in order, at the end of the log; any other set records the last summary as before. Shard and count lines that scroll out of the 64 KiB output tail are kept, so a long shard run still sums.CI_CARRY: none (… tree_mismatch …)now prints the one-command fix:gh workflow run <workflow> --ref <branch>, then gstack-ci-carry wait(TRUT-7).workflow_dispatchrun reports under the same job names on the head commit, so without strict branch protection a green dispatch run can hide a red merge-commit check.tree_changedrecheck.Why
gbrain's sharded lane recorded about half its tests, so a carried lane under-reported coverage. TRUT-7 hit the bare
tree_mismatchrefusal on a real pull request and had to work out the dispatch fix by hand.Blast radius
bin/gstack-evidence(test-count parsing),lib/run-evidence.tsandlib/ci-carry.ts(refusal text only: no refusal becomes a carry), and docs. No network calls, secrets, permissions, workflows or dependencies change.Test Coverage
test/evidence.test.ts: two--shard=i/2processes in one log record 6095, two shards with more than 64 KiB of per-test lines record 4000, and an incomplete, repeated or rerun shard set records the last summary. The sum cases failed before the change.test/ship-ci-carry.test.ts: a pull_request run whose tree.json records a real merge commit and its tree prints the base-moved reason and the dispatch command. It failed before 39b1c59 with the baretree_mismatchthe real run printed.Pre-Landing Review
REVIEW_LOOP: passes=2 blocking_found=2 deferred=0 noted=3 fixed=4 scope=frozen unrecorded=0 ledgers_read=4 fix_checks=2 dispatches=3
Light depth (code class). Pass 1 found one CRITICAL (shard blocks scrolled out of the 64 KiB tail, so the sum silently fell back) plus two Low notes. All three were fixed in 65ae8d4. Pass 2 found nothing. After the merge with main, a standing fix-check on the docs audit's edit found one CRITICAL. The base-moved hint sat on a check real runs never reach, because CI's tree.json records the merge commit's tree, so the earlier
rec.treecheck refused first with no hint. 39b1c59 moves the hint onto that refusal for pull_request runs whose recorded sha's tree is the recorded tree. A scoped fix-check on that hunk found nothing at the floor.Review-loop exception: accepted by GStack Minder 2026-11-03 via GSTA-258 (GSTA-36 choice A, #3934 merge-tree hint): attach the base-moved hint to the :213 tree_mismatch refusal when it's a pull_request run whose rec.sha tree equals rec.tree; replace the fixture with a failing-first one that records the merge tree; one scoped fix-check on that hunk as #3934's second exception, no further rounds (fix-check at 39b1c59, 0 findings)
Open notes (not fixed in this PR):
autoCarryin lib/pregate/index.ts cuts theCI_CARRY: nonereason at 300 characters after joining up to 10 candidate refusals, so a second earlier refusal or a long branch name can push thefix:text past the cut."tree": nullwith an unreadable merge sha gets the hint on itstree_mismatch; it still refuses.Issues
GBEV-3 and TRUT-7 item 3 (GSTA-258 L6 follow-up).
Refs #3923
Refs #3929
Refs #3934
Documentation
Status: current. The re-audit after l6fdocs3 found the docs accurate at 39b1c59, so no edits were made.
Scope: committed diff d2e6318..39b1c59 (bin/gstack-evidence, lib/ci-carry.ts, lib/run-evidence.ts, docs/pregate.md, changelog.d fragment, two tests); no staged, unstaged or new files. Focus: commit 39b1c59's merge-tree
tree_mismatchfix hint, docs/pregate.md #ci-carry, #dispatch-carry and #skipped-files, and any doc describingtests_ranas the last summary only.Documentation health:
docs/pregate.md Current (the #ci-carry sentence matches verifyRun: only the tested-merge-commit refusals carry
fix: gh workflow run free-tests.yml --ref <branch>, then gstack-ci-carry wait; other tree_mismatch mentions, #dispatch-carry and #skipped-files are accurate)CONTRIBUTING.md Current (CI carry row accurate; no tests_ran counting claim)
docs/standalone-binaries.md Current (no tests_ran or tree_mismatch claim)
docs/troubleshooting.md Current (no tests_ran or tree_mismatch claim)
docs/designs/HONEST_WORK_GATE_INTEGRITY.md Current as a dated design record (states last-summary-wins without the GBEV-3 shard sum; not edited)
changelog.d/gsta258-l6-carry-followup.md Read only (tree_mismatch line slightly narrower than code; see decisions)
Coverage debt: the
tree_mismatchfix hint and the bun shard-set sum have reference coverage only (docs/pregate.md, changelog fragment), with no how-to. Diagram drift: none.GSTA-258 condition met: audit l6fdocs3 read the whole merged docs/pregate.md, both #3911's flaky-exception section and #3929's sections, and found them current apart from one edit.
Test plan
LANE: free carried run=41662215702/1 tests_ran=1560 (files)gstack-pregateverify ok on the stamped tree; remote gate cleared from run 41662215702 (skipped=70), macOS lane from the same runtscclean