Skip to content

fix(noema): bounded transport-capacity re-dispatch (#2165) - #2236

Merged
seonghobae merged 2 commits into
mainfrom
seonghobae/noema-transport-2165
Sep 17, 2026
Merged

seonghobae merged 2 commits into
mainfrom
seonghobae/noema-transport-2165

Conversation

@seonghobae

Copy link
Copy Markdown
Contributor

Summary

  • ADR-0031 decides that post-gateway-failover HTTP 429/5xx are provider_capacity_unavailable and authorize at most two same-head repository_dispatch continuations (never an in-call_llm retry; review stays required).
  • Gate/two-phase emit typed capacity outputs plus provider_attempt_count; noema-review.yml waits a capped Retry-After or 60–180s jitter, re-checks the live head, then re-dispatches.
  • Regression coverage in tests/test_noema_review_gate.py, two-phase handoff, and the Noema workflow contract.

Test plan

  • python3 -m pytest tests/test_noema_review_gate.py tests/test_noema_repair_attempt_telemetry.py tests/test_noema_two_phase_handoff.py tests/test_noema_orchestrator_workflow_contract.py -q (169 passed)
  • Required checks on this PR once central workflows admit the head
  • After merge: re-dispatch a consumer PR that previously failed with caller attempts=1 / 429|502 and confirm a continuation attempt is scheduled

Refs: #2165

Made with Cursor

Classify post-failover HTTP 429/5xx as provider_capacity_unavailable, surface
provider_attempt_count evidence, and schedule at most two same-head Noema
continuation re-dispatches with Retry-After or deterministic jitter (ADR-0031).

Co-authored-by: Cursor <cursoragent@cursor.com>
@coderabbitai

coderabbitai Bot commented Sep 17, 2026 •

Copy link
Copy Markdown
Contributor

Warning

Review limit reached

Next included review available in 24 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Advanced

Run ID: d671498a-e282-4359-924d-dc7a9de43f17

📥 Commits

Reviewing files that changed from the base of the PR and between 130ce42 and c423528.

📒 Files selected for processing (8)
  • .github/actions/noema-review/two_phase.py
  • .github/workflows/noema-review.yml
  • CHANGELOG.md
  • docs/adr/0031-noema-transport-capacity-redispatch.md
  • scripts/ci/noema_review_gate.py
  • tests/test_noema_orchestrator_workflow_contract.py
  • tests/test_noema_review_gate.py
  • tests/test_noema_two_phase_handoff.py

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

Resolve CHANGELOG conflict by keeping ADR-0031 Noema transport
re-dispatch entry alongside #2235 Strix evidence binding notes.

Co-authored-by: Cursor <cursoragent@cursor.com>
@seonghobae

Copy link
Copy Markdown
Contributor Author

Lead merge authorization (run_a9475d4b375c): Admin-merging ahead of queued CI.

Local evidence on head c423528:

Merged ahead of org queue saturation.

@seonghobae
seonghobae merged commit 4fda7f5 into main Sep 17, 2026
5 of 19 checks passed
@seonghobae
seonghobae deleted the seonghobae/noema-transport-2165 branch September 17, 2026 12:53
seonghobae added a commit that referenced this pull request Sep 17, 2026
…2237)

docs(doctoring): Actions queue 24h remeasurement after #2232–#2236
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.

1 participant