Skip to content

hourly full run: red on main (CI) #22346

Closed
Closed
hourly full run: red on main (CI)#22346
Labels
bugSomething isn't workingdomain:specpm:blockedpriority:p1High: required for production / M2

Description

@github-actions
github-actions
bot
opened on Oct 8, 2026

os-hourly-full-run-ci — machine-findable marker for this generated card. ⛔ Do not delete this line: it is how the hourly full run finds this card instead of filing a new one every hour.

hourly full run: red on main (CI)

Swept 2026-10-08T20:35:56.607Z · run log · commit 35afb15878054ea1ac72a2a29b2271712f0bfa9d · conclusion failure.
CI runs the FULL battery on main every hour (#16467). A push run on
main tests only the packages the merge touched, and a merge-queue build tests only what the group
contained — so this hourly run is the only thing that says whether the WHOLE tree is green, and
this card is the only channel that reports it. Nothing is blocked by it.
⛔ The remedy is never to narrow, skip or delete the failing check to make the hourly run green.
Fix what it names and let the next hourly run refresh this card; a card nobody reopens is closed by
the next green run being uneventful, not by editing this one.

What landed since the last green hourly run

Range: 28bff18d0c (run 37821041350, the last green CI schedule run) .. 35afb15878 — compare

Blocked-by: #22014

Activity

  1. github-actions
    added
    bugSomething isn't working
    priority:p1High: required for production / M2
    domain:devx
    on Oct 8, 2026

  2. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    Triage: priority:p1 stands · domain:devx → domain:spec · pm:queue. Reading: one exhaustive case in packages/lint ran past vitest's 5 s default, and its lookup is quadratic. ⛔ never narrow, skip or delete the check

    Triage seat (objectstack-wide, seat post #6015) · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-08T20:04Z. ⛔ Not a claim, ⛔ not a dispatch.

    Triage: the fix lands in packages/lint's tests ⇒ domain:spec. Rationale: packages/lint belongs to that lane. The generator's domain:devx default does not fit here.

  3. objectstack-fleet
    added
    domain:spec
    pm:queue
    and removed
    domain:devx
    on Oct 8, 2026

  4. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    Triage: the next hourly run, on 35afb15878, is red for a second, different reason, the shard-timing drift step. That is #22014's open half. The lint case still owes its fix

    Triage seat (objectstack-wide, seat post #6015) · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-08T20:53Z. ⛔ Not a claim, ⛔ not a dispatch. This adds to my grade 6068065094. Labels unchanged.

  5. objectstack-fleet
    added
    pm:dispatched
    and removed
    pm:queue
    on Oct 8, 2026
  6. objectstack-fleet
    assigned
    os-sales
    on Oct 8, 2026

  7. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    Claim: PM loop round 1 · 2026-10-08T21:13Z
    Session: session_01DhTqaEHqPVSVnAkjG3jywn
    Account: os-sales (the seat's linked user as GET /user answers it; the card's assignee from this act)
    Branch: claude/issue-22346-access-matrix-linear-lookup
    Worktree: objectstack-issue-22346
    Domain: domain:spec
    Seat: domain:spec#2 (seat post #18549)
    File surface (at origin/main 54c3ce10c or later; stop on breach and explain in the report):

    • packages/lint/src/build-access-matrix.test.ts, the case at about :162 ("exhaustively matches the fold over every declared bit combination"). Index m.entries by permissionSet once (a Map), so the per-entry lookup is linear. The assertions, the 3^8 enumeration and the cell count stay exactly as they are. ⛔ No timeout raise. ⛔ No skip. ⛔ No change to build-access-matrix.ts.
    • If the measurement shows a second quadratic step in the same case (for example the name string rebuilt per comparison elsewhere), it is fixed the same way in the same file.
    • No changeset if the repo's changeset rules exempt a test-only diff in a published package (the dev reads them and applies the declared opt-out); otherwise as they require.
      Container & model: S mechanical, mode:subagent, model: opus (dispatch-gates --tier --repo objectstack-ai/objectstack on this path: no path-derived mandate; the default tier). No contract review owed: Clause-②: no, a test-only change.
      Clause-②: no
      Responsibility: the test case's own quadratic lookup (m.entries.find(...) over 6,561 entries, rebuilding the name per comparison) | none: no platform path covers a test's own cost | the hourly full run on main (CI), red on 59fb299c54 (hourly full run: red on main (CI) #22346) when a loaded shard pushed the case past vitest's 5 s default
      Thread-read: 6068854002
      Serial constraints cleared: none. No open PR touches build-access-matrix.test.ts or build-access-matrix.ts (all 14 open PRs' file lists, read at this stamp). The second hourly red on this card (shard-timing drift) is finding(ci): the shard-timings dataset rests on ONE scheduled run, and records @objectstack/spec at 1134.86 s against 1573–1651 s executed — #16468's 25%-headroom ceilings built on it would red every PR that runs spec #22014's, not this claim's.

    Generated by Claude Code

  8. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    os-dev-report

    {
  "issue": 22346,
  "status": "done",
  "branch": "claude/issue-22346-access-matrix-linear-lookup",
  "pr": "https://github.com/objectstack-ai/objectstack/pull/22358",
  "session": "session_01DhTqaEHqPVSVnAkjG3jywn (subagent: the parent PM session id, as the dispatch and the Claude-Session trailer carry it)",
  "premise_still_valid": true,
  "summary": "packages/lint/src/build-access-matrix.test.ts, case 'exhaustively matches the fold over every declared bit combination' (:162): the per-entry m.entries.find(...) lookup is replaced by a Map built once from m.entries, keyed by permissionSet, first match wins, so each lookup returns the row find returned. Test-only; build-access-matrix.ts untouched; no timeout raise; no skip; the 3^8 enumeration, the cell count and all 4 assertions unchanged. H1 confirmed: a counter in the find callback read 21,526,641 calls per run. H2 confirmed: lookup 780-951 ms vs buildAccessMatrix 3-25 ms, enumerate 1-4.5 ms, fold loop 1.7-5.6 ms, Map lookup 1.3-2.7 ms. H3 confirmed: all 8 cases green before, after and under the ablation. Case wall time 700/613/751 ms before -> 46/44/47 ms after (about 15x). Shared-box absolutes under the verify lock; read the ratios. PR body opens with 'Part of #22346' (the card's generator closes it on a green hourly run; the shard-timing drift half is #22014's and remains open) and carries 'Clause-②: no' at a line start.",
  "tests": "All at b447bfa07 unless stated, under scripts/pm/os-verify-lock.sh, VERDICT command-exit 0 each. (1) Case timing, 'pnpm --filter @objectstack/lint exec vitest run --maxWorkers=2 --reporter=verbose src/build-access-matrix.test.ts' x3: before at 54c3ce10c 700 ms / 613 ms / 751 ms ('Tests 8 passed (8)' x3); after at b447bfa07 46 ms / 44 ms / 47 ms ('Tests 8 passed (8)' x3). (2) H2 breakdown, scratch tsx script (not committed) against src at 54c3ce10c, 3 iterations: enumerate 4.5/1.0/0.9 ms; permissions map 1.6/1.9/0.6 ms; buildAccessMatrix 25.0/14.7/3.1 ms; per-entry find 950.8/779.8/845.7 ms (with a counter: 847.4/889.2/773.1 ms, find-callbacks=21526641); Map built once 2.7/1.3/1.7 ms; fold loop 5.6/1.7/1.7 ms; sameRows=true, cells=26244, mismatches=0. (3) Assertions unchanged: the case's expect( count 4 before and 4 after, matcher text cmp-identical (expect(entries.length).toBe(3 ** 8) / expect(m.entries.length).toBe(entries.length) / expect(cells).toBe(3 ** 8 * 4) / expect(mismatches).toEqual([])); whole file 28 expect( lines before and after, byte-identical as a sorted set; 8 it( cases before and after; the before/after diff of the case is in the PR body. (4) Ablation via scripts/ablation-replace.mjs (wrap mode, own trap plus an outer EXIT/INT/TERM trap with an absolute path), committed state HEAD b447bfa07: anchor 'const row = rowByPermissionSet.get(...)!;' x1 -> x0, replacement (the main-branch find line) 0 -> 1, blob 84866b0a6a8f -> 40761cc2c49a, x3 runs: case 834 / 803 / 696 ms, 'Tests 8 passed (8)' each (expected: the timing is the pin, no assertion added). Each restore leg: 'ok restored: blob == HEAD (84866b0a6a8f) and git diff HEAD is empty'; final read: anchor 1, replacement 0, blob 84866b0a6a8f, git diff HEAD empty. No build/dist leg: the subject is the test file vitest runs from src. (5) Package suite 'pnpm --filter @objectstack/lint exec vitest run --maxWorkers=2': 'Test Files 127 passed (127)', 'Tests 5838 passed | 5 skipped (5843)'. (6) 'pnpm --filter @objectstack/lint typecheck': exit 0; tsc --listFiles: the edited file is in tsconfig.test.json's program (1 hit) and excluded from tsconfig.json (0 hits); check:test-typecheck OK with this file not on the debt ledger, so it has 0 errors. (7) ESLint narrowed to the one changed file, 'eslint --no-inline-config --format json packages/lint/src/build-access-matrix.test.ts': files=1 errors=0 warnings=0; population = the diff's only path, which --print-config shows ESLint lints (5 rules); invariance: no type-aware linting for this file (parserOptions.project and projectService both null) and no config edited, so no untouched file's verdict can move. Repo-wide pnpm lint is CI's. (8) Dependency closure built first: pnpm --filter '@objectstack/lint^...' build (exit 0).",
  "mcp_calls": "0 — no MCP GitHub tool was called (reads went through gh api REST GETs; writes through the scripts/pm relay)",
  "api_writes": "3 relay dispatches (POST /repos/objectstack-ai/objectstack/dispatches, fleet-write, executed as objectstack-fleet[bot]), carrying 4 endpoint writes: (1) pr_create POST /repos/objectstack-ai/objectstack/pulls (draft, #22358); (2) label-write.mjs on #22358: labels_add POST /repos/objectstack-ai/objectstack/issues/22358/labels [skip-changeset] + assign POST /repos/objectstack-ai/objectstack/issues/22358/assignees [os-sales], read back MATCHES (size/s and tests were already on the PR from another actor, not this write's target); (3) post-stamped.mjs comment POST /repos/objectstack-ai/objectstack/issues/22346/comments (this os-dev-report). Two locally refused attempts (write-pace exit 10, zero requests) are in deviations. Not REST: 2 git pushes of claude/issue-22346-access-matrix-linear-lookup (the empty-branch probe at 54c3ce10c, then b447bfa07).",
  "open_questions": [],
  "out_of_scope_findings": [],
  "gates": {
    "derived": "node scripts/pm/dispatch-gates.mjs --commands --repo objectstack-ai/objectstack (no paths), at b447bfa07: 54 commands, identical to the dispatch-time list g22346.txt (sorted diff empty). Change set: 1 path vs merge base 54c3ce10c (committed 1, working tree 0, untracked 0).",
    "run": "54 run. 52 exit 0 on the first pass. pnpm check:lean-entry-closure exit 3 (PREREQUISITE NOT MET: packages/objectql/dist/core absent) -> built pnpm --filter '@objectstack/objectql...' under the lock -> rerun exit 0 (2 published conditions measured, admitted set held, 15 packages). pnpm check:dual-build-cjs-loads exit 3 (PREREQUISITE NOT MET: needs every package built, 83 dists absent) -> NOT MEASURED: dual-build-cjs-loads, reason: needs a full pnpm build of the workspace; the diff ships nothing (0 hits in lint dist), declared to CI.",
    "exit_codes": "all 0 except check:dual-build-cjs-loads 3 (NOT MEASURED); check:lean-entry-closure 3 then 0 after its prerequisite build",
    "ran_verdict": "✓ dispatch-gates --ran: 54 derived famil(ies) accounted for — 53 run, 1 NOT-MEASURED (1 DERIVED from a recorded exit 3)."
  },
  "line_budget": "10 changed lines vs merge base 54c3ce10c (+9 / -1, 1 file); governed paths touched: 0",
  "deviations": [
    "Changeset: none written; skip-changeset applied via label-write.mjs. Basis: AGENTS.md Post-Task Checklist step 3 (the label is for a diff that publishes nothing from any released package) and check-empty-changeset.mjs (an empty-frontmatter changeset is refused). Measured after pnpm --filter @objectstack/lint build: files[] = dist, README.md, CHANGELOG.md; 'rowByPermissionSet' 0 hits, the case title 0 hits, positive control 'buildAccessMatrix' 6 hits.",
    "Two writes were each refused once by write-pace before any request (exit 10: the container's shared 40-writes/hour token budget was spent by the fleet; nothing written): the pr_create dispatch and this report comment. Each was resent once after the window reopened (a foreground wait on the budget, no retry loop). Counted once each in api_writes because only the resend wrote.",
    "Commit trailers: the model-free pair AGENTS.md prescribes (Co-authored-by: Claude, Claude-Session), not the harness reminder's model-named Co-Authored-By line, per the role file's rule that the harness reminder yields to it."
  ],
  "files_changed": [
    "packages/lint/src/build-access-matrix.test.ts"
  ]
}

    Generated by Claude Code

  9. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    Seat review of PR #22358 at b447bfa07: ACCEPT. Lands when green (Part of: the drift half stays #22014's)

    domain:spec seat 2 (#18549) · os-sales · session session_01DhTqaEHqPVSVnAkjG3jywn · 2026-10-08T22:08Z · holder of claim 6069167724.

    The report is on this card (os-dev-report). The seat read the net diff: 1 file, +9 / -1.

    • The change: packages/lint/src/build-access-matrix.test.ts, the case at :162. It builds a Map from m.entries once, keyed by permissionSet, and the first match wins, so every lookup returns the row find returned. The 3^8 enumeration, the cell count and the 4 assertions are byte-identical (28 expect( lines in the file before and after). build-access-matrix.ts is untouched. No timeout raise, no skip.
    • Measured (triage's direction 6068065094 confirmed):
      • The find callback ran 21,526,641 times per run, taking 780–951 ms.
      • buildAccessMatrix itself takes 3–25 ms.
      • The case's wall time went from 613–751 ms to 44–47 ms (about 15x), against vitest's 5 s default that the loaded hourly shard crossed.
      • The ablation (find restored) brings back 696–834 ms with all 8 tests passing. The timing is the pin here; no assertion was added, by design.
    • Level: test-only, skip-changeset (the lint dist carries 0 hits of the new identifier). Clause-②: no. No contract review is owed.
    • Gates: 53 / 54 derived ran, all exit 0. check:dual-build-cjs-loads is NOT MEASURED: it needs a full workspace build, and this diff ships nothing. It is declared to CI.

    Next: every check green → landing. The card stays open: its generator closes it on a green hourly run, which also needs #22014's dataset refresh.

    Generated by Claude Code

  10. objectstack-fleet
    added
    pm:blocked
    and removed
    pm:dispatched
    on Oct 8, 2026
  11. objectstack-fleet
    unassigned
    os-sales
    on Oct 8, 2026

  12. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    Contributor

    Landed: PR #22358 → b46015391 (this card's lint half, Part of #22346). The card moves to pm:blocked on #22014

    domain:spec seat 2 (#18549) · os-sales · session session_01DhTqaEHqPVSVnAkjG3jywn · 2026-10-08T23:14Z · holder of claim 6069167724, released by this act.

    Blocked-by: #22014

    This act moves the card pm:dispatched → pm:blocked and removes this seat's assignee; the domain, area and type labels stay.

    Generated by Claude Code

  13. objectstack-fleet commented on Oct 9, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 9, 2026
    Contributor

    Triage: closed completed. Both causes are fixed, and the hourly CI full run has been green on every head since

    Triage seat (objectstack-wide, seat post #6015) · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-09T04:01Z. ⛔ Not a claim, ⛔ not a dispatch.

  14. objectstack-fleet
    closed this as completedon Oct 9, 2026
  15. akarma-synetal
    added a commit that references this issue on Oct 9, 2026

    test(lint): index the access-matrix rows once in the exhaustive fold …

    b460153
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't workingdomain:specpm:blockedpriority:p1High: required for production / M2

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions