[kixelated]

Problem

A moq-transport subscriber cannot bound the age of groups the publisher serves. IETF delivery timeouts also expire the newest group and cannot represent moq-lite's subscription budget.

Closes #5881.

Approach

Negotiate subscriber max age in SETUP, carry the existing max-delay budget on SUBSCRIBE and REQUEST_UPDATE, and apply it through the existing group-expiration policy in Rust and JavaScript. Relays retain the most tolerant subscriber's budget; absence preserves retained backlog, and the newest group survives every budget. Metadata query subscriptions follow the aggregate too, so an initial budget cannot mask a later decrease. Update the Timestamp draft and standard guidance.

Direct subscription fixtures now explicitly declare their peer SETUP. Delayed-SETUP tests and peers without max-age support retain their existing negotiation assertions.

Impact

• Public API: none.
• Wire, alongside MoQ Epoch: SUBSCRIBE may carry both EPOCH (0x40B69, draft 17+) and MAX_AGE (0x915C6); keys are encoded in ascending order.
• Wire: additive MAX_AGE Setup Option 0x915C4 and subscriber parameter 0x915C6, in milliseconds, on drafts 14 through 22. Send the parameter only to peers declaring the option.
• Delivery timeouts remain accepted and ignored.

Alternatives

Reusing OBJECT_DELIVERY_TIMEOUT or SUBGROUP_DELIVERY_TIMEOUT would cut newest groups, including quiet catalog snapshots. Sending an unnegotiated parameter would close legacy sessions.

Validation

Validated commit 7116ef3dfe8dce247c2d43e5a0f954e09cac9778:

• just check: passed, including all 7,535 native tests, additional feature suites, workspace JavaScript tests and type checks, documentation, lint, and target checks.
• just drafts check: passed.
• just rs test -p moq-net: all 2,518 tests passed, 8 skipped. The initial broad run exposed two joining-FETCH fixtures that never supplied peer SETUP; explicit fixture declarations fix both without changing production negotiation.
• just rs test -p moq-net -E 'test(subscriber_max_age) | test(ietf::subscribe::tests) | test(partial_group::)': all 49 tests passed, including most-tolerant aggregation and budget decreases across supported drafts, newest catalog survival, and partial-group resets.
• 13 targeted JavaScript regressions passed. The original publisher fails a requested 250ms budget as unlimited. The active-reader relay regression fails without the query-hold update: the publisher stays at 250ms instead of the requested 10ms after a tolerant subscriber leaves.
• just test interop --publishers rust,python,go,cpp --subscribers rust,python,go,cpp,js-native-node,js-native-bun,c,gst --timeout 30: passed all available native matrix, unauthorized-publisher, CLI FETCH, and finite-tail checks.
• just test drill ietf_subscriber_max_age_skips_bufferbloat: passed. The drill uses a finite 500ms queue below the media bitrate and verifies overflow drops, with random loss disabled.
• just test drill-sensitivity --no-baseline subscriber-max-age-ignored: passed, 1 of 1. Removing initial publisher max-age application makes the drill fail with subscriber max age did not reach the publisher. Patch application also passed separately.

Iteration (Claude Opus 5.5)

Merged origin/main (MoQ Epoch, #5836 and neighbors). Conflicts were unions of the two extensions' fields in Peer, PeerSetup, Subscribe, the SETUP parameter enum, and the JS subscriber. The Rust parameter macros require ascending keys, so EPOCH now precedes MAX_AGE in SUBSCRIBE encode and decode. No review findings were posted.

Decisions: none needed; the merge adopted main's PeerSetup::new and kept this PR's PeerSetup::declared test fixture.

Validated on the merge commit:

• just check: passed.
• just drafts check: passed.
• just rs test -p moq-net (nextest, all features): 2,572 passed, 8 skipped.
• js/net typecheck and bun test src/ietf: 701 passed.
• just test interop --all, browser included: all checks passed. This closes the gate the previous iteration left open.

Round 2 (Claude Opus 5.5)

Codex finding r4320231271: JS draft-17+ dial and accept dropped maxAge when destructuring the SETUP result, so JS never sent MAX_AGE on ALPN-selected drafts. Both call sites now spread the whole result. A new js/net/src/integration.test.ts case covers drafts 16, 17, and 22 end to end; 17 and 22 fail without the fix. just check and just test interop --all pass on the fix.

Merged origin/main again (refresh-ratio dynamic groups): SUBSCRIBE and REQUEST_UPDATE keep the max age alongside main's fixed-range bounds and join-judged NEW_GROUP_REQUEST, and the subscriber-max-age-ignored drill mutation is retargeted (the drill passes at baseline and fails with the mutation). just check and moq-net and js/net tests pass on the merge; just test interop --all was rerun on the final head.

Follow-ups

None.

🤖 Generated with Claude Code

(Written by GPT-6.1-Sol; iterated by Claude Opus 5.5)

[coderabbitai]

Warning

Review limit reached

You'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 11 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used all 4 included reviews currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 88411873-36a4-44dc-a33e-5a9192088b35

📥 Commits

Reviewing files that changed from the base of the PR and between 3b77d23 and b684e8d.

📒 Files selected for processing: 37

• doc/concept/standard.md
• drafts/draft-lcurley-moq-timestamp.md
• js/net/src/connection/accept.ts
• js/net/src/connection/connect.ts
• js/net/src/connection/handshake.ts
• js/net/src/ietf/connection.ts
• js/net/src/ietf/index.ts
• js/net/src/ietf/max-age.test.ts
• js/net/src/ietf/max_age.ts
• js/net/src/ietf/parameters.ts
• js/net/src/ietf/publisher.test.ts
• js/net/src/ietf/publisher.ts
• js/net/src/ietf/subscribe.ts
• js/net/src/ietf/subscriber.test.ts
• js/net/src/ietf/subscriber.ts
• js/net/src/integration.test.ts
• rs/moq-net/src/client.rs
• rs/moq-net/src/fuzz.rs
• rs/moq-net/src/ietf/adapter.rs
• rs/moq-net/src/ietf/max_age.rs
• rs/moq-net/src/ietf/mod.rs
• rs/moq-net/src/ietf/parameters.rs
• rs/moq-net/src/ietf/peer.rs
• rs/moq-net/src/ietf/publisher.rs
• rs/moq-net/src/ietf/request_stream.rs
• rs/moq-net/src/ietf/session.rs
• rs/moq-net/src/ietf/subscribe.rs
• rs/moq-net/src/ietf/subscriber.rs
• rs/moq-net/src/ietf/version.rs
• rs/moq-net/src/model/origin.rs
• rs/moq-net/src/server.rs
• rs/moq-net/tests/integration/main.rs
• rs/moq-net/tests/integration/partial_group.rs
• rs/moq-net/tests/integration/subscriber_max_age.rs
• rs/moq-relay/tests/drills.rs
• test/drill/README.md
• test/drill/mutations/subscriber-max-age-ignored.patch

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-20T02:07:34.421165Z	b684e8d	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 9b506f9259

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".