Issue metadata
Sign in to add a comment
|
Chrome webRTC does not control the send bandwidth based on b=AS and b=TIAS proposed by remote SDP (video codec: H.264)
Reported by
yinbao2...@gmail.com,
Mar 21 2018
|
||||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.162 Safari/537.36 Steps to reproduce the problem: webRTC client running on Chrome (developed using JSSIP) receives an incoming call with SDP specifying the following: - session BW: 256kbps - media audio BW: 64 kbps - media video BW: 128 kbps Below is setRemoteDescription copied from chrome://web-internals: type: offer, sdp: v=0 o=- 554044268 554044268 IN IP4 172.21.1.100 s=Asterisk c=IN IP4 172.21.1.100 b=AS:256 t=0 0 m=audio 10174 RTP/SAVPF 0 101 b=TIAS:64000 a=connection:new a=setup:actpass a=fingerprint:SHA-1 69:E8:5C:65:E4:E0:8F:32:9A:65:57:5D:FF:8D:30:93:5D:F2:8E:FA a=ice-ufrag:4a0597123bf8bc2b06c15a17474ff9d2 a=ice-pwd:54e7001c63946625132138e12bd2a39e a=candidate:Hac150164 1 UDP 2130706431 172.21.1.100 10174 typ host a=candidate:S22c5e729 1 UDP 1694498815 34.197.231.41 10174 typ srflx raddr 172.21.1.100 rport 10174 a=candidate:Hac150164 2 UDP 2130706430 172.21.1.100 10175 typ host a=candidate:S22c5e729 2 UDP 1694498814 34.197.231.41 10175 typ srflx raddr 172.21.1.100 rport 10175 a=rtpmap:0 PCMU/8000 a=rtpmap:101 telephone-event/8000 a=fmtp:101 0-16 a=ptime:20 a=maxptime:150 a=sendrecv a=rtcp-mux m=video 16222 RTP/SAVPF 99 b=TIAS:128000 a=connection:new a=setup:actpass a=fingerprint:SHA-1 69:E8:5C:65:E4:E0:8F:32:9A:65:57:5D:FF:8D:30:93:5D:F2:8E:FA a=ice-ufrag:7f7858002990484a656978c35bf19269 a=ice-pwd:01bd38c9107bf289070edf6620ace39d a=candidate:Hac150164 1 UDP 2130706431 172.21.1.100 16222 typ host a=candidate:S22c5e729 1 UDP 1694498815 34.197.231.41 16222 typ srflx raddr 172.21.1.100 rport 16222 a=candidate:Hac150164 2 UDP 2130706430 172.21.1.100 16223 typ host a=candidate:S22c5e729 2 UDP 1694498814 34.197.231.41 16223 typ srflx raddr 172.21.1.100 rport 16223 a=rtpmap:99 H264/90000 a=fmtp:99 max-mbps=108000;packetization-mode=1;profile-level-id=42000B a=sendrecv a=rtcp-mux What is the expected behavior? webRTC should not send video frames with bit rate higher than what's specified in remote SDP. I expect the googTransmitBitate/googActualBitrate to be within 128-256kbps What went wrong? actual googTransmitBitate/googActualBitrate is observed to be around 1.5Mbps, where googAvailableSendBandwidth is observed as high as 1.5-4Mbps. I have attached the event log and webrtc-internal-dump. Did this work before? N/A Does this work in other browsers? N/A Chrome version: 65.0.3325.146 (Official Build) (64-bit) Channel: n/a OS Version: Mac OS X Flash Version: The attached log is a short call. I have also ran the same on Win10 based Chrome and observed the same: even on long calls the send BW is around 1 - 1.5Mbps. If there is another way I should use to control the BW used by webRTC app please let me know. Our downlink device is a mobile device that does poorly when flooded with high BW video traffic. Thanks!
,
Mar 22 2018
,
Apr 2 2018
I have not seen recent update on this issue. Can you please let me know if there's any additional info needed? Thanks.
,
Apr 3 2018
The session-level b=AS isn't implemented, and media-level b=TIAS isn't implemented either, so every bandwidth limit in your SDP is being ignored. Here's a fiddle that can be used to test this: https://jsfiddle.net/emson7ap Also, b=AS has some other issues... See https://bugs.chromium.org/p/chromium/issues/detail?id=803615
,
Apr 9 2018
This may be a deviation from the original issue: are rtfp-fb supported by Chrome? Specifically the following 3: a=rtcp-fb:* nack pli a=rtcp-fb:* ccm tmmbr a=rtcp-fb:* ccm fir I have a low throughput downstream device and would like to configure webRTC in Chrome to either limit the BW or support flow control. Appreciate any suggestion, thanks.
,
Apr 9 2018
PLI and FIR are supported and advertised in generated SDP. TMMBR appears to be supported, but doesn't appear in SDP, so maybe its implementation has caveats. holmer@, can you summarize the status of TMMBR? Why don't we advertise support for it? Here's a related bug: https://bugs.chromium.org/p/webrtc/issues/detail?id=7103
,
May 21 2018
Going ahead and closing this as duplicate of bugs.webrtc.org/5788 (b=TIAS not implemented) and bugs.webrtc.org/9296 (b=AS not implemented at session level). |
|||||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||||
Comment 1 by vamshi.kommuri@chromium.org
, Mar 22 2018Labels: Triaged-ET Needs-Triage-M65 TE-NeedsTriageHelp