Issue metadata
Sign in to add a comment
|
WebRTC - network-id on onIceCandidate since Chrome 51,
Reported by
pascal.b...@gmail.com,
Jun 4 2016
|
||||||||||||||||||||||
Issue descriptionUserAgent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.79 Safari/537.36 Steps to reproduce the problem: I am working in ALE International (formerly Alcatel-Lucent Enterprise). In OpenTouch solution, we propose anonymous clients to connect/join meeting with "OTC Web" which allows to join audio with WebTRC (audio codec G722). Architecture for WebRTC: browser (Chrome or Firefox) -> Internet -> SBC -> Opentouch Server (on DMZ) 1. No problem with Chrome 50 2. Since Chrome 51 : ICEConnectionStateFailed What is the expected behavior? Acts like Chrome < 51 or give me the RFC (didn't find it) which describes this network-id. I need this to open a support request to our SBC provider. What went wrong? WebRTC negotiation failed with ICEConnectionStateFailed I noticed by comparing SDP negotiation side by side that since Chrome 51, on onIceCandidate: sdpMid: audio, sdpMLineIndex: 0, candidate: candidate:3266345979 1 udp 2122255103 2001::5ef5:79fd:10f8:12b1:a756:9f61 63052 typ host generation 0 ufrag cN4WNsfYhohiix0O network-id 1 there is a "network-id X" at the end of ALL onIceCandidate. Not reproduced by downgrading to Chrome 50. This is also not on Firefox. Is it a RFC ? I found a ref here for Chromium: Issue 1815473002: Add 16-bit network id to the candidate signaling. (Closed) https://codereview.webrtc.org/1815473002/#ps80001 Regards, Pascal Did this work before? N/A Chrome version: 51.0.2704.79 Channel: stable OS Version: 10.0 Flash Version: Shockwave Flash 21.0 r0
,
Jun 6 2016
,
Jun 6 2016
Assigning to the author of https://codereview.webrtc.org/1815473002
,
Jun 6 2016
There is an IETF draft for it: https://tools.ietf.org/html/draft-thatcher-ice-network-cost-00 But it has not been accepted yet by the ICE Working Group. But if you're parsing the candidate line correctly, there shouldn't be any problem with seeing "network-id 1", just like there shouldn't be a problem with "ufrag XXX" or "generation 0". If you don't understand one of these extra attributes, just ignore it.
,
Jun 6 2016
Ok thanks, I'll transfer this to Audiocodes (provider of our SBC)...
,
Jun 14 2016
I'm also facing same problem. did you resolve this?
,
Jun 14 2016
Hi, @work, I transferred the question/infos to Audiocodes and they provide us a new SBC to test: Version F7.00A.067.003 After deploying it in our test Lab all went well with Chrome. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by pascal.b...@gmail.com
, Jun 4 2016