Latest version of Chrome not including `a=msid`

367 views
Skip to first unread message

se...@pion.ly

unread,
Apr 1, 2024, 12:10:16 PM4/1/24
to discuss-webrtc
I have been getting a lot of bug reports into Pion about this.

Does anyone have a crbug, was this a planned deprecation? 

Chrome Media Section
```
m=audio 9 UDP/TLS/RTP/SAVPF 111 9 0 8
c=IN IP4 0.0.0.0
a=rtcp:9 IN IP4 0.0.0.0
a=ice-ufrag:MfuH
a=ice-pwd:fig1ADQ0ADTtEMWH77NptwLz
a=ice-options:trickle
a=fingerprint:sha-256 36:D4:A4:44:0A:D5:97:8E:D4:6F:32:80:2A:87:F1:59:D1:8C:10:53:03:B1:4F:44:D7:4B:10:0E:2E:AA:CE:64
a=setup:active
a=mid:1
a=extmap:1 http://www.ietf.org/id/draft-holmer-rmcat-transport-wide-cc-extensions-01
a=sendonly
a=rtcp-mux
a=rtpmap:111 opus/48000/2
a=rtcp-fb:111 transport-cc
a=fmtp:111 minptime=10;useinbandfec=1
a=rtpmap:9 G722/8000
a=rtpmap:0 PCMU/8000
a=rtpmap:8 PCMA/8000
a=ssrc:352663822 cname:uljllCW8U9c39XGw
```

se...@pion.ly

unread,
Apr 1, 2024, 12:23:31 PM4/1/24
to discuss-webrtc
This seems to only happen when Offer is a `recvonly` transceiver. I will have a jsfiddle shortly that shows the behavior difference

se...@pion.ly

unread,
Apr 1, 2024, 12:50:08 PM4/1/24
to discuss-webrtc
Here is a jsfiddle that exhibits the behavior https://jsfiddle.net/6jrypdez/9/

se...@pion.ly

unread,
Apr 1, 2024, 1:08:46 PM4/1/24
to discuss-webrtc
I found the relevant crbug[0]

Adding `a=msid-semantic: WMS*` causes Chrome to respond with a `a=ssrc *msid*` line now.

Philipp Hancke

unread,
Apr 1, 2024, 2:02:12 PM4/1/24
to discuss...@googlegroups.com
If you still rely on the plan-b ssrc specific msids you should stop doing that now

--
This list falls under the WebRTC Code of Conduct - https://webrtc.org/support/code-of-conduct.
---
You received this message because you are subscribed to the Google Groups "discuss-webrtc" group.
To unsubscribe from this group and stop receiving emails from it, send an email to discuss-webrt...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/discuss-webrtc/32d22273-945e-40c2-a709-2bab35dd8545n%40googlegroups.com.

Sean DuBois

unread,
Apr 1, 2024, 2:16:30 PM4/1/24
to 'Neil Young' via discuss-webrtc
Thanks Neil! What values should I be using?


The following values are in the example unified plan offer

```

a=ssrc:11 cname:cname

a=ssrc:11 msid: track-id-2

a=ssrc:11 mslabel:

a=ssrc:11 label:track-id-2

```

How should I be determining the MediaStreamTrackId instead?

Philipp Hancke

unread,
Apr 1, 2024, 11:37:28 PM4/1/24
to discuss...@googlegroups.com, Harald Alvestrand
That document  is pretty outdated (as any doc not containing a name). RFC 8829 recommends looking for a=msid:stream (no space there) but ignores that due to backward compat with existing software it is advisable to include the track id too. And OFC does not consider that there is no way to distinguish that from "legacy" software that does not include msid at all...

Harald: can you do a semi-hostile doc takeover

Reply all
Reply to author
Forward
0 new messages