Skip to content

ci: raise Merkle E2E timeout to 90 min — 60-min cap breached by real runtime (V2-1046) - #178

Merged
grumbach merged 1 commit into
mainfrom
nic/v2-1046-merkle-e2e-timeout
Aug 25, 2026
Merged

grumbach merged 1 commit into
mainfrom
nic/v2-1046-merkle-e2e-timeout

Conversation

@Nic-dorman

Copy link
Copy Markdown
Member

Linear issue

V2-1046

Risk tier

  • T0 — docs / tooling / CI / pure UX-output. Repo CI only.
  • T1 — client-only, no network-facing behavior change. CI + prod compat smoke.
  • T2 — node/client logic with behavioral surface, no protocol/format/economics change. Dev testnet + ADR.
  • T3 — protocol / storage format / payments / routing. T2 evidence + adversarial testing.

Compatibility

  • Wire: none
  • Storage: none
  • API: none

Semver impact

  • breaking
  • feature
  • fix

Test evidence

Evidence the cap is breached by healthy runs, from PR #174's run (an additive API change touching nothing merkle-related): macOS leg passed at 57m09s; Ubuntu leg canceled at 60m20s with no test failure in the log — the workflow's timeout-minutes: 60 killed it mid-run. Third breach of this cap (20 → 40 → 60; history in the workflow comment, now updated). This PR's own Merkle E2E jobs exercise the new bound.

New dependency

none

ADR

n/a

Mitigation / rollback

CI-only — revert the single commit; worst case a genuinely hung job holds a runner 30 min longer before the kill.

🤖 Generated with Claude Code

…runtime (V2-1046)

Third time this cap has been outgrown (20 -> 40 -> 60): on PR #174 the
macOS leg passed at 57m09s and the Ubuntu leg was canceled at 60m20s
mid-run with no test failure in the log — a pure timeout kill reading
as a red X on an unrelated change. 90 min gives the observed ~60-min
runtime real headroom while still catching genuine hangs; the
neighboring E2E job has no explicit cap, so this stays the tighter
bound.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@dirvine

dirvine commented Aug 21, 2026

Copy link
Copy Markdown
Member

Review verdict: APPROVE at 68d852abfcfd41c79c929cd6cf129a35c3372ebd.

This is a narrowly scoped CI-only change: Ubuntu Merkle E2E gets 60 minutes while macOS stays at 45, matching the observed runner asymmetry without weakening the test itself. The workflow expression and self-tests are sound; local verification of .github/scripts/test_check_pr.py passed 9/9, and both Python files compile. I found no diff-specific blocker.

The current Security Audit failure is the separate repository-wide h2 advisory, not caused by this timeout change.

@grumbach
grumbach merged commit a954ec8 into main Aug 25, 2026
14 of 15 checks passed
grumbach added a commit to grumbach/ant-client that referenced this pull request Aug 25, 2026
Picks up the Merkle E2E timeout raise from WithAutonomi#178 (60 -> 90 min), which is
what killed the ubuntu leg of this branch's last CI run at 60m18s with no
test failure. Also brings in the beta upgrade channel work from WithAutonomi#173.

No conflicts; the branch's own diff against main is byte-for-byte
unchanged by this merge.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants