fix(net): skip old groups without reporting them as lost

[kixelated]

A group superseded by a newer one is skipped by design (the default max_age of zero), but two paths logged that skip at WARN as if data were lost:

• resume::Group::give_up warned "no route can serve the rest of this group" when the route's copy died with Old.
• frame::Raw's drop fallback warned "frame::Producer dropped before writing all bytes" even when the group was already aborted (superseded, evicted, cancelled) and carried its own reason. The lite group receiver also dropped its ingest before aborting the group, so a track or group closing mid-frame hit the same warning.

Found on moq.pro production: its stats aggregator reads a few thousand one-frame-per-tick .stats tracks with the default budget. It logged about 50 of these warnings per second (46k lines in 15 min), and each relay serving it logged several per second, for groups that were skipped exactly as intended.

API / wire impact: none. Readers get the same errors as before (lost_group_stays_lost still covers Old); only the log level of an expected skip changes. There's no regression test because moq-net has no log-capture harness, and the change is log-only.

Not addressed here: web_transport_moq::session: failed to decode unidirectional stream err=WebTransportError(UnknownSession) shows up at a similar rate on the same hosts.

🤖 Generated with Claude Code

(written by Claude Opus 5.5)

[coderabbitai]

Warning

Review limit reached

Next included review available in 2 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used all 4 included reviews 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: 34501ec6-887e-4973-89c7-e1fd7b9e0e94

📥 Commits

Reviewing files that changed from the base of the PR and between be5404d and 8104f1d.

📒 Files selected for processing (3)

• rs/moq-net/src/lite/subscriber.rs
• rs/moq-net/src/model/frame.rs
• rs/moq-net/src/model/resume.rs

---

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	⚠️ Failed 2026-09-25T23:07:43.890918Z	8104f1d	Draft marked ready

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.

[kixelated]

Local just check passes on this branch with #4210 applied (4327 + 297 + 107 tests). Without it, the gate fails only on the moq-video capture-test compile error on macOS that #4210 fixes.

(written by Claude Opus 5.5)

[kixelated]

Landing on main. A group skipped because a newer one superseded it stays a skip: give_up no longer warns on StreamError::Old, and dropping an in-flight frame into an already-aborted group no longer warns. Reader errors are unchanged. Check and Test are green.

(Written by Grok 4.7)