Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 0 additions & 4 deletions js/net/src/ietf/subscriber.ts
Original file line number Diff line number Diff line change
Expand Up @@ -2006,10 +2006,6 @@ export class Subscriber {
// the track ends with it, and at object 0 it ends before it.
if (producer && frame.idDelta !== 0) throw new ObjectIdGap(frame.idDelta);
await last();
// Our own unsubscribe can release the track while its marker is in flight. A
// released track has no end left to declare, and the publisher did nothing wrong,
// so refusing it would close the session and every other subscription with it.
if (hooks.trackReleased(track).peek() !== undefined && !hooks.trackFinished(track)) return;
const end = { group: group.groupId, frame: nextObject + frame.idDelta };
producer?.finish();
try {
Expand Down
6 changes: 4 additions & 2 deletions js/net/src/track.ts
Original file line number Diff line number Diff line change
Expand Up @@ -2318,13 +2318,15 @@ export class Producer {
// The wire's end: repeating it is a no-op, and a mid-group end may refine the group
// boundary it falls just before. Either must carry the same error code. Mirrors the Rust
// `finish_at` on a `received()` copy. Only a wire subscription owes a tail below the end: a
// FETCH's end leaves nothing in flight, as the Rust `set_end` a fill declares.
// FETCH's end leaves nothing in flight, as the Rust `set_end` a fill declares. An end that
// lands after the track closed, such as a viewer releasing it while the marker was in
// flight, has nothing left to end.
#endPending(end: Position, error?: Error): void {
const declared = this.#state.end.peek();
const code = (err?: Error) => (err === undefined ? undefined : toStreamCode(err));
const tail = this.#wire !== undefined;
if (declared === undefined) {
if (this.#state.closed.peek() !== undefined) throw new Error("track is closed");
if (this.#state.closed.peek() !== undefined) return;
this.#declareEnd(end, error, tail);
} else if (code(declared.error) !== code(error)) {
throw new ProtocolViolation(`conflicting track end error: ${error}`);
Expand Down
4 changes: 3 additions & 1 deletion quest/m0/interop-green/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -212,7 +212,9 @@ Scheduled runs
The JS subscriber refused the END_OF_TRACK of a withdrawn route after its own
unsubscribe had released the track, and a protocol violation closes the whole
session.
[moq-dev/moq#8769](https://github.com/moq-dev/moq/pull/8769) tolerates the marker on a released track.
[moq-dev/moq#8769](https://github.com/moq-dev/moq/pull/8769) tolerated the marker on a released track,
and [moq-dev/moq#8754](https://github.com/moq-dev/moq/pull/8754) moved that into the track's wire end, so a
late PUBLISH_DONE or FETCH_OK keeps the session too.
That is not the whole case: since [moq-dev/moq#8443](https://github.com/moq-dev/moq/pull/8443)
dropped the Chromium quarantine, `just test media --media-cases replacement`
at `ce616b52c` failed 4 of 14 local runs (host at load 50), as `live incumbent
Expand Down
4 changes: 4 additions & 0 deletions quest/m0/nightly/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -57,4 +57,8 @@ report's output ([moq-dev/moq#8654](https://github.com/moq-dev/moq/pull/8654)),
the firefox resume ([moq-dev/moq#8635](https://github.com/moq-dev/moq/pull/8635)),
and the chromium `av_lag` rows
([moq-dev/moq#8676](https://github.com/moq-dev/moq/pull/8676)).
After #8635 landed, run
[43601796176](https://github.com/moq-dev/moq/actions/runs/43601796176) still
failed the firefox resume, now on audio arriving late rather than trailing:
[Firefox resume audio](/quest/m0/nightly/firefox-resume-audio.md) owns it.

36 changes: 36 additions & 0 deletions quest/m0/nightly/firefox-resume-audio.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,36 @@
# [M] Firefox audio resumes with the picture

## Goal

Nightly's `Media (firefox player)` job passes its `after resume` window: after
a pause, audio is audible within the window's first samples, as in Chromium,
instead of staying silent for about half a second after the picture resumes.

## Plan

[moq-dev/moq#8635](https://github.com/moq-dev/moq/pull/8635) removed the
1.7 s trail behind the picture, but Nightly
[43601796176](https://github.com/moq-dev/moq/actions/runs/43601796176) (at
4fec611d4, after it and the [lag rule](https://github.com/moq-dev/moq/pull/8722))
still failed with `68/79 samples carried the fixture tone` (11-13: 60/78). The
window's first 11 samples are silent with the audio byte counter flat
(`audio=94041B`) while video presents live frames; then about 1.8 KB of
audio lands in one burst, `[player] skipping missing groups: track=audio
1165 -> 1258` fires, and audio plays in sync. So the audio subscription's
first groups reach the Firefox player about 0.4 s after video's.

Not reproduced locally: `just test media --media-engines firefox
--media-cases rename,pause` passed 4 of 4 on an idle host (2026-11-14). The
CI host was 14 to 60% idle during the window, and Firefox dials the relay
over WebSocket, so the difference is likely in that path under load.

Recommended: reproduce under load first (the job's own
`busiest host processes` table shows the shape), with the relay's and
player's logs kept (`HARNESS_KEEP=1`), and find which hop holds audio's
first groups after the resubscribe: the relay's upstream re-subscribe and
gap backfill (the fixture logs a fetch of audio group 1156 that the relay
aborts), the WebSocket session's stream scheduling, or the consumer's hole
wait from its pre-pause cursor (1165). Fix that hop; do not widen the test's
window or agreement.

Kept in m0 by the maintainer (2026-11-15): ✅ m0 (rejected: m1, drop).