Repository navigation
ci: raise the Ubuntu test shard timeout to 20 minutes - #8944
Conversation
Hosted ubuntu-latest shards were cancelled at the 15-minute wall in their last batches with no failing tests (#8651 shard 1 at batch 62/65 and shard 5 at 66/66; also #8585, #8645, #8841, #8861, #8887). A cancelled shard is neither a pass nor a fail, so it blocked merges without evidence of a fault. Raise only the test job's timeout-minutes to 20 and pin the new ci.yml blob with an appended release-evidence policy entry. A shard far beyond 20 minutes is still treated as wedged.
Codex Review Summary: ✅ CompletedThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
No actionable comments were generated in the recent review. Walkthrough
Merge Risk:
|
|
✅ Deterministic PR hygiene checks passed. |
|
Maintainer integration record: implements owner decision decision:8b360a80 (raise the Ubuntu test shard cap). Squash-merged on exact head 19a262d with all 31 required checks green, security review approved on this head (only the test job's timeout-minutes changed; parsed workflow otherwise identical; release-evidence policy additions-only with the ci.yml pin matching the tree), and |
Summary
ci.ymljobtest,test N/5) keep getting cancelled at their 15-minutetimeout-minuteson hostedubuntu-latestrunners while they are in their last batch and have no failing tests. feat(combos): opt-in capabilityPolicy "union" advertises member capabilities and filters per request #8651 lost shard 1 at batch 62/65 and shard 5 at batch 66/66, and the same happened on fix(combos): reserve output headroom and margin estimates in target admission #8585, fix(combos): cool targets after committed output_limit/stall and hop commentary-only degenerate completions #8645, fix(responses): reject repeated Chat work announcements #8841, feat(jev): hide the effort control on a decide-mode JEV Combo #8861 and fix(responses): keep display estimates out of Combo spend accounting #8887. A cancelled shard is neither a pass nor a fail, so each one blocked a merge without showing any fault.testjob'stimeout-minutesfrom 15 to 20. Its comment now explains the hosted-runner headroom and still says that a shard running far beyond the cap is wedged. No other job's timeout changes, and triggers, runners and CI policy text stay as they are.tests/ci-workflows/ci-workflows.test.tsnow expects 20 fortest;gatesstays at 15. The comment above it no longer claims the Windows shards share the Linux budget..github/scripts/release-evidence-policy.jsongets one newcientry at the end, copied from the latest one. The only changed pin is.github/workflows/ci.yml→4ae44e193c6813a0378700f37c1cebaca7327cf1(git hash-object --no-filters). This is a pure addition, and earlier entries are untouched.structure/or docs page states the 15-minute Ubuntu shard cap (checked withrg), so no docs change is needed.Verification
Run locally on
19a262d08a(parentca96ead01a, currentorigin/dev):node .github/scripts/release-evidence.cjs policy-check: "Evidence policy fingerprints match the working tree."node --test .github/scripts/release-evidence.test.cjs: 26 pass, 0 failbun run typecheck: no type errors (2,633 files)bun run structure:check: passedbun run privacy:scan: passedgit diff --check: cleanbun test --isolate --timeout 60000 tests/ci-workflows/ci-workflows.test.ts tests/ci-workflows/ci-structure-gate.test.ts tests/ci-workflows/file-size-ratchet.test.ts tests/ci-workflows/docs-spanish-localization.test.ts tests/ci-workflows/structure-ssot.test.ts tests/test-layout.test.ts tests/test-layout-tooling.test.ts: 283 pass, 0 failbun test --isolate --timeout 60000 tests/ci-workflows: 2033 pass, 27 skip (Windows-only Authenticode cases), 0 fail across 102 filesNOT RUN: full
bun run test. The change is limited to one workflow timeout, its contract test and a policy pin, and the wholetests/ci-workflowsdomain that covers them was run above. The workflow change itself needs the exact-head PR CI.Checklist
Summary by CodeRabbit