[finding] driver-sql's live PG + MySQL matrix runs under vitest's default 5000ms — the only live-DB driver in the repo with no testTimeout, and it just dequeued an unrelated PR #16434

Activity

1. zhuangjianguo

   added

   findingRecorded observation — held, not dispatchable until the findings triage round grades itRecorded observation — held, not dispatchable until the findings triage round grades it

   on Sep 7, 2026
2. claude commented on Sep 7, 2026

   

   claudeBot

   on Sep 7, 2026 – with Claude

   Contributor

   More actions

   Second instance, in a different package — and it dequeued a second unrelated PR

   domain:engine dispatching seat, 2026-09-07T05:5xZ. ⛔ Still an observation, no grading, no domain:*, no priority — this is added because it widens the card's population, not because it changes its verdict.

   The new instance

   Queue CI run 34086147182, Test Core (4/6) — 17 jobs, exactly one failing test:

   FAIL  packages/plugins/plugin-email/src/email-service.queue-delivery.test.ts:279
     > EmailService — queue delivery on
     > still refuses the queue for attachments OVER the limit, and stores nothing (#5177)
       ↳ Error: Test timed out in 5000ms.
   

   It dequeued #16442, whose eight changed files include zero under packages/plugins/plugin-email/, and whose own PR CI is 41/41 with 0 failing on the same head.

   The same probe, the same answer, a different package

   probe	driver-sql (this card, 05:4xZ)	plugin-email (new)
   testTimeout in the package's vitest.config.ts	absent	absent
   explicit per-test timeouts in the failing file	0	0
   effective budget	vitest default 5000 ms	vitest default 5000 ms
   what it cost	dequeued #16430 (a changeset-body-only PR)	dequeued #16442 (a @objectstack/core narrowing)

   The control table from the original filing is unchanged and still fires: six packages in this repo do set testTimeout — driver-mongodb 30_000, service-datasource 60_000, spec 60_000, qa/http-conformance 30_000, plugin-auth 10_000, metadata-fs 10_000.

   ⚠️ What this does and does not change about the card

   Does: the population is wider than "the one live-DB driver". Two packages, two queue-only timeouts, two dequeued PRs, in one night — and neither failing test belongs to the PR it dequeued. Whatever the fix is, scoping it to driver-sql alone would leave the second instance standing.

   Does not: it does not make the two instances the same defect. driver-sql's is specific and derivable — 24 live round-trips to two real servers inside a 5 s budget. plugin-email's failing test is in-process and has no live server, so why it needs more than 5 s is a separate question this seat has not answered, and ⛔ raising its budget without answering it would convert a visible flake into a slow test nobody notices. That question belongs to whoever takes this card.

   ⚠️ ⛔ And the direction NOT to take by reflex: a repo-wide default is the tempting fix and it is the one that costs the most — it raises the ceiling for every fast in-process suite, which is where a 5 s guard is doing real work. The two derivations (what a live-server matrix needs; whether an in-process email test should need it at all) are separate, and the second one may end in "fix the test" rather than "raise the budget".

   ---

   Generated by Claude Code
3. os-zhuang

   added

   tests

   toolingFix lands in tooling, not a product package — enters pm:queue only with Unblocks: or a named surfaceFix lands in tooling, not a product package — enters pm:queue only with Unblocks: or a named surface

   priority:p2Medium: important, M3Medium: important, M3

   pm:queueReady for the PM dispatch loopReady for the PM dispatch loop

   domain:engineRetired 2026-10-22, alias of domain:server (engine surface); in-flight cards keep it until they landRetired 2026-10-22, alias of domain:server (engine surface); in-flight cards keep it until they land

   and removed

   findingRecorded observation — held, not dispatchable until the findings triage round grades itRecorded observation — held, not dispatchable until the findings triage round grades it

   on Sep 7, 2026
4. os-zhuang

   added the

   Bug

   issue type on Sep 7, 2026
5. os-zhuang commented on Sep 7, 2026

   

   os-zhuang

   on Sep 7, 2026 – with Claude

   Contributor

   More actions

   Triage: lands in domain:engine (packages/drivers/driver-sql); rationale: class (a) — ⭐ this one did not predict a cost, it measured one that was already paid. PR #16430 (+46 / −1 on a single .changeset/*.md, zero files under driver-sql/) was dequeued at 00:45:57Z with CI_FAILURE on a test its diff cannot reach, and the queue rebuilt everything behind it. Bug.

   priority:p2. Two facts set it there and both are on the card: the failure mode is a merge-queue dequeue, which is strictly more expensive than a red PR check, and the 24h ledger recorded 0 other queue failures — so this is the first instance, which is exactly when the budget question is cheap to answer rather than tangled with a backlog of re-runs.

   ⭐ The firing control is what makes this gradeable rather than arguable, and it should survive into the PR: six packages set testTimeout; driver-mongodb — the other live-DB driver — sets 30_000; driver-sql, the one package whose suite talks to two live servers, sets none and inherits vitest's 5000ms. A complete grep, a named comparator, and an explicit statement of what the finding does not claim (5s is fine for the in-memory / SQLite cells, which are the majority). That last fence is the difference between a budget card and a "raise the timeout" reflex.

   ⭐ This is a missing budget, not a flaky assertion, and the card proves it rather than asserting it: §2 deliberately does no sleeping — the file's own head docblock records that #11572 kept "§1, §1b and §2 … with no sleep between them" because the unslept same-millisecond path is §2's subject. What it spends is network: ROUNDS = 6 × four live round-trips = 24 live round-trips in one test, with the cursor comparison deliberately made server-side so it happens in the column's own type. ⇒ ⛔ Do not "fix" §2 by sleeping, backdating or driving its clock. The two prior repairs to this file (#11572 drove §3 instead of racing it; #13691 moved §3's validity ceiling) touch §3 only and ⛔ must not be extended to §2 — doing so would delete the thing the test measures.

   ⛔ Do not treat the next occurrence as a flake and re-run it. A re-run was the right response to this instance and it was already spent (#16430 re-queued at 00:49:01Z). The card is about the class, and the cause is named.

   Route — the two questions the card says a dev must settle rather than assume, ratified here as blocking, not optional:

   1. One budget or two. A package-wide testTimeout also raises the ceiling for the fast in-memory cells, weakening them as a guard; a live-cells-only budget keeps them strict at the cost of being carried explicitly. ⇒ Decide and say which, in the PR.
   2. What number, derived from what. ⛔ Not copied from driver-mongodb by analogy — 30_000 is that package's number, not a repo constant. Derive it from the observed live round-trip cost, the discipline [finding] sql-driver-11224-update-stamp-precision §3 now reds on a SLOW runner too — the span < 1000ms line is a validity precondition, and #11572's fix spends deliberate clock time inside that same budget #13691 applied to §3's span ceiling: a bound derived from the conclusion it guards, not picked. An undated, underived timeout is indistinguishable from one raised to silence a real regression.

   Sibling note: #16497 (core's discriminator pin, same default 5000ms) is graded p3 today — same class, but observed only on a shared dev box and never on CI. This card outranks it because the cost here is on the record.

   ⛔ This seat grades and routes only: not claimed, not dispatched, no code.

   ---

   Generated by Claude Code
6. akarma-synetal

   added a commit that references this issue on Sep 28, 2026

   test(core): charge the entry-point pin's index import to collect, not…

   ...

   24d0542