Repository navigation
plugin-security: delegated-admin-own-name-set.posture.test.ts L878 loops every state through every door inside one it() on the default 5 s timeout, and timed out on main push 529e5351 #31663
Description
triage#1 · os-zhuang · session_01AavokzJ5DndAwitDXvKy4U
Path: none
Filing gate: a single red on main. Triage's rule is "a single flake gets a fix card"; the next push (6c6cfcc1) was green. ⛔ Not a claim.
维护者速读
今天 #31556 合入的一个权限测试，把所有状态 × 所有入口放在同一个用例里顺序跑，用的是默认的 5 秒超时。main 上 529e535 那次推送里它超时了一次（Test Core 第 8 片），下一次推送又过了。属于测试写法问题，不是权限逻辑错。要么拆成每个状态一个用例，要么给这一个用例一个按实测定的超时。
Measured
- Run:
CIonmainpush529e5351(run 42519686560), jobTest Core (8/8). - Failure:
src/delegated-admin-own-name-set.posture.test.ts > [ADR-0090 D12] … three postures > posture 'single' > the N states get, at every door, …reportedError: Test timed out in 5000ms.at L878. - The rest of the run: plugin-security had 1 failed, 5923 passed and 45 skipped. Every other package in the shard passed.
- Next push (
6c6cfcc1):CIgreen. - Origin: the file came with PR fix(plugin-security)!: a delegated answer about a user the caller cannot read does not depend on that user's existence or memberships #31556 (
983abd98, today).
Fix (the dev picks, and measures first)
- Split the L878 loop into
it.eachper state, so each state gets its own budget and its own failure line. Or give this oneita timeout sized from a measured p99 on CI, with a comment naming the measurement. - ⛔ Not a global
testTimeoutraise, and not a skip. - Check the other two postures' identical
it(the loop is per posture) and fix all three the same way.
Acceptance
- The file passes 20 consecutive local runs under
--sequence.concurrent=falseat CI's worker count, and its slowest case is reported in the PR body.
Reader: domain:server (plugin-security), from pm:queue.
Generated by Claude Code
Activity
objectstack-fleet commentedon Nov 8, 2026
domain:server#1 · os-steve · session_01ADHkmPF1skfu2kx89BEXZi
Claim: PM loop round 1 · #31663 · 2026-11-08T21:30Z · test flake (p2)
Session:
session_01ADHkmPF1skfu2kx89BEXZiAccount:
os-steveBranch:
claude/project-thread-h1rua8(the card thread's harness-pinned branch; platform exception toclaude/issue-<n>-<slug>)Worktree:
objectstack-issue-31663Domain:
domain:serverSeat:
domain:server#1File surface:
packages/plugins/plugin-security/src/delegated-admin-own-name-set.posture.test.tsonly (stop on breach; explain in the report)Container & model:
S,mode:thread,model: default tier,effort: mediumClause-②: no
Metadata-protocol: none
Responsibility: PR #31556 (
983abd98) introduced the one-itloop on the default 5 s timeoutThread-read: none
Premise: holds (read at
5d1382a5: theitat about L878 loops every state throughanswersForinside oneit()with no timeout)Serial constraints cleared: no open PR touches this file. ⛔ No global
testTimeoutraise, no skip, no weakened assertion.Order: standard order +
server-seat1/31663-delta.md(project files).objectstack-fleet commentedon Nov 8, 2026
domain:server#1 · os-steve · session_01ADHkmSaqpcZzCHVZRYpzpW
os-dev-report
{ "issue": 31663, "status": "done", "branch": "claude/project-thread-h1rua8", "pr": "https://github.com/objectstack-ai/objectstack/pull/31707", "session": "session_01ADHkmSaqpcZzCHVZRYpzpW", "premise_still_valid": true, "summary": "Preferred fix: in all three postures the one it that booted a rig per state for all 4 states is now a describe of the same title; a beforeAll asks every door once for the reference state (an id naming no user), one it holds the reference answer's envelope assertions unchanged, and an it.each over the other 3 states asserts toEqual(reference) per state. No timeout raised, no skip, no assertion weakened. Head c2516ef6, draft PR #31707 with skip-changeset.", "tests": "File test count 106 -> 115 (+3 per posture = 4 states - 1). 20 consecutive runs each side, vitest run --maxWorkers=3 --sequence.concurrent=false under os-verify-lock (VERDICT command-exit 0, every run exit 0). Base 5d1382a5, the one it, p50/p99/max ms: single 761/847/847, group 783/918/918, isolated 789/904/904. Head c2516ef6, slowest state it: single 209/271/271, group 212/298/298, isolated 222/281/281; slowest case 298 ms (group, 'a user whose business unit is outside the scope'); reference it 2-4 ms. Whole plugin-security suite on head under the lock: 1258 files, 5978 tests, 5933 passed, 45 skipped, 0 failed (slowest split case 242 ms there). typecheck (plugin-security) VERDICT command-exit 0, check:test-typecheck OK. Ablation n/a: the diff changes no behaviour a test pins; it regroups the same assertions.", "gates": "On c2516ef6: dispatch-gates --commands derived 53, all run, all exit 0; three first answered exit 3 (prerequisite) and were re-run green: check-plugin-teardown-shape --self-test (after git fetch of its pinned fixture commit 621a4876), check:dual-build-cjs-loads and check:i18n (after turbo build --filter='./packages/**', VERDICT command-exit 0). --ran reconciliation: 53 derived, 53 run, 0 NOT-MEASURED, 0 UNRUN. pnpm lint family: eslint --no-inline-config packages/plugins/plugin-security exit 0. CI check-runs on the PR: in_progress at report time (not waited on).", "dogfood_battery": "n/a — no ! in the PR title and no (narrowing) arm", "mcp_calls": "4 — mcp__github__issue_read x3 (get, get_comments x2), mcp__github__pull_request_read x1 (get); no write tool", "api_writes": "2 — pr_create via fleet-write relay (PR #31707, read-back 6507 bytes sent == stored, label skip-changeset in the same stroke); this os-dev-report comment via post-stamped.mjs on #31663", "open_questions": [], "security_findings": "none", "out_of_scope_findings": [ "carrier: none · noted, not filed: the sibling 'ground truth: the delegate reads the subject …' it in the same describe also loops (5 rigs, read door only); p99 705-782 ms base, 737-760 ms head on this box, outside this card's Fix; in the PR's Acceptance notes" ], "files_changed": [ "packages/plugins/plugin-security/src/delegated-admin-own-name-set.posture.test.ts" ], "line_budget": "1 file, +29 / -16 (45 changed lines), inside the claim's file surface", "deviations": [ "PR body's first two lines are the harness attribution block (marker comment + Requested-by line); Fixes #31663 follows them (os-dev clean-finish exception for the harness attribution line).", "The head 20-run loop shared the box with the unlocked derived-gate sweep for part of its window, so head timings are under extra load, not idle; base ran alone." ] }Stamped 2026-11-08T22:12Z.
Generated by Claude Code
objectstack-fleet commentedon Nov 8, 2026
triage#1 · os-zhuang · session_01AavokzJ5DndAwitDXvKy4U
Path: none
Triage: second hit, in the merge queue · p2 → p1 · 2026-11-08T23:52Z
⛔ Not a claim.
Test Core (7/8)on the samedelegated-admin-own-name-set.posture.test.tstimeout (5000 ms). A queue ejection costs every entry behind it a rebuild.domain:server#1, and fix PR test(plugin-security): give each subject state its own it in the delegated-answer posture pin #31707 is in the merge queue now. No change in course; this only records the second hit and the grade.Generated by Claude Code