Repository navigation
quest: the JS namespace lifecycle cell waits on an @moq/net patch - #6503
Conversation
Quest-Claim: c8e12e0678ffadff
…rage/js-namespace-done
|
Outcome: the cell fails because (Written by Claude Opus 5.5) |
# Conflicts: # quest/m1/interop-coverage/README.md # quest/m1/interop-coverage/js-namespace-done.md # quest/m1/interop-coverage/runner-cases.md
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
Warning Review limit reachedYou've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Next included review available in 33 minutes. View limit details
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. Comment |
Codex Review SummaryThis 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. |
|
Merging. The maintainer decided to ship #6518 JS-only on release, and this quest now waits on the @moq/net 0.4.6+ patch that carries it. CI is green, the current head has a non-Grok review, and (Written by Claude Opus 5.5) |
# Conflicts: # quest/m1/interop-coverage/README.md
|
Merging. Requeued after merging main; the only change since review is that merge. CI is green, the current head has a non-Grok review, and (Written by Claude Opus 5.5) |
|
Merge triage: quest-only change (2 files), (Written by Sonnet 5.5) |
Finds why the interop runner's moq-dev-js
subscribe-namespace-lifecyclecase never sees NAMESPACE_DONE against Cloudflare draft-18 and aiomoqt (englishm/moq-interop-runner#173), fixes it onrelease, and rewrites the quest to wait on the next@moq/netpatch release that carries the fix, then the runner pin bump.Cause
Not shared state between two sessions in one process. The published
@moq/net0.4.5 and earlier withdraw a PUBLISH_NAMESPACE on drafts 17+ with FIN plus STOP_SENDING. Draft-18 §3.3.2 cancels a request by terminating every open direction abruptly. Both relays end the namespace only on RESET_STREAM, so the subscriber waits until the publisher's session closes. From two processes the case passed only because the exiting publisher closed its session.Reproduced with the runner's client on the
releasesource, trying each signal alone: FIN, STOP_SENDING, FIN then STOP_SENDING after the ack, and FIN plus STOP_SENDING all fail on both relays; RESET_STREAM alone passes. Our sender was out of spec, so nothing goes upstream.mainalready sends the reset since moq-dev/moq#5483, and the runner's client built onmainpasses against both relays. The published Rust client (moq-tokio0.19.28) passes too, over WebTransport and raw QUIC.Changes
js/nethalf of fix(net): withdraw PUBLISH_NAMESPACE by reset on draft-17+ #5483 torelease, with its tests, plus a subscriber fix so aCANCELLEDreset ends the request without an error log. It merged as 350dee5. With it the runner's client passes against Cloudflare d18, aiomoqt d22, imquic, moxygen, cdn.moq.dev, and stitcher-moq.@moq/netpatch release (0.4.6 or later) shipping fromreleasewith fix(net): withdraw PUBLISH_NAMESPACE by reset on draft-17+ (release backport) #6518; checknpm view @moq/net versionis above 0.4.5 and thatrelease'sjs/net/package.jsonbump to it lands after 350dee5. Then bump the runner's moq-dev-js client to it. It also updates the quest's line in the epic. (The runner-cases quest it used to edit was retired onmainby moq-dev/moq#6505; the merge takes that deletion.)Public API and wire impact
CANCELLED) instead of FIN + STOP_SENDING, as onmain.Checks
quest checkandjust checkpass on this branch after mergingmain.just check origin/releasepasses. The three changed publisher tests and the new subscriber test fail without the fix. CI passed and Codex approved its final head.just test interop --allpasses except the browserjscells, which could not launch Chromium (release's Playwright against the dev shell's browsers). That harness runs moq-lite, which the change does not touch.Decisions
releasefor the next@moq/netpatch (maintainer). The 0.4.5 bump moq-dev/moq#6501 had already shipped without it, so the quest waits on 0.4.6 or later. No version bump here; releases are cut separately.releaseRust also FINs on 17+, but the published Rust client passes the same cells against both relays.(Written by Claude Opus 5.5)
🤖 Generated with Claude Code