1917909 - RTX SSRC set to zero after re-negotiation

Status: UNCONFIRMED

(Core :: WebRTC, defect, P3)

Description

[Alexander Kuznetsov]

Steps to reproduce:

1. Server sends offer SDP with one m=video sendonly line and rtx payloads and rtx ssrc
2. Firefox negotiates SDP successfully
3. RTX working correctly
4. Firefox call AddTrasnceiver("video") and create local offer SDP with two m=video lines - first line from previous server SDP, and second line is local m=video sendonly line. And send SDP to server.
5. Firefox negotiates SDP successfully
6. RTX not working.

In the Firefox logs I see that the rtx_ssrc is 0 for fist m-line.
I did a little research.
If some parameters in SDP changes, WebrtcVideoConduit::OnControlConfigChange() called - https://searchfox.org/mozilla-central/source/dom/media/webrtc/libwebrtcglue/VideoConduit.cpp#482

This always creates a new webrtc::VideoReceiveStreamInterface::Config::Rtp - https://searchfox.org/mozilla-central/source/dom/media/webrtc/libwebrtcglue/VideoConduit.cpp#511 with rtx_ssrc = 0

rtx_ssrc sets if remoteSsrcUpdateNeeded is not false - https://searchfox.org/mozilla-central/source/dom/media/webrtc/libwebrtcglue/VideoConduit.cpp#880

But remoteSsrcUpdateNeeded is ALWAYS false, because the SSRC values ​​have not changed - https://searchfox.org/mozilla-central/source/dom/media/webrtc/libwebrtcglue/VideoConduit.cpp#496

Actual results:

rtx not working after second SDP negotiation

Expected results:

rtx working correctly

Comment 1

[Alexander Kuznetsov]

Maybe it is same bug?... https://bugzilla.mozilla.org/show_bug.cgi?id=1894997

Comment 2

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Core::WebRTC' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 3

[Michael Froman [:mjf]]

Byron, does this look like a dup of Bug 1894997?

Comment 4

[Byron Campen [:bwc]]

I'm pretty sure it is. I have a work in progress on this, but unfortunately we do not have a way to test it in automation. Since we have a report from the field, I think we should go ahead and fix without having a test.

Comment 5

[Michael Froman [:mjf]]

Since this bug is from the field, I'm going to match the priority/severity from Bug 1894997 and then dup that bug to here.