- From: Harald Alvestrand <harald@alvestrand.no>
- Date: Tue, 07 Apr 2015 21:00:19 +0100
- To: public-webrtc@w3.org
- Message-ID: <55243753.8090406@alvestrand.no>
On 04/07/2015 06:24 PM, Roman Shpount wrote: > On Tue, Apr 7, 2015 at 1:15 PM, Byron Campen <docfaraday@gmail.com > <mailto:docfaraday@gmail.com>> wrote: > > On 4/7/15 9:56 AM, Roman Shpount wrote: >> On Tue, Apr 7, 2015 at 12:42 PM, Byron Campen >> <docfaraday@gmail.com <mailto:docfaraday@gmail.com>> wrote: >> >> The current jsep spec requires the c-line to have "IN IP6 >> ::" in it until candidates are gathered, and once candidates >> are gathered it is set to a real address. Are you dealing >> with an implementation that supports trickle ICE, but is >> picky about the contents of the c-line? >> >> Firefox uses '0.0.0.0' because it is slightly more >> interoperable than '::', but probably won't do so indefinitely. >> >> >> I am dealing with implementations that do not support trickle ICE >> and getting the SDP after ICEGatheringStateComplete. I am >> expecting to see a real IPv4 address in the "c=" line at that point. > Ok, that bug was fixed in 37, so you should be good to go now. > > > Yes, I can confirm that 0.0.0.0 address in the "c=" line produced by > Firefox is fixed in 37. > > Chrome still produces IPv6 addresses in the "c=" line in some random, > hard to predict circumstances, so this SDP muck is still there. >From Byron's comment it seems that there are cases where IPv6 will be produced by conforming implementations. What's the situation under which IPv6 in the c= line will cause problems? > _____________ > Roman Shpount > -- Surveillance is pervasive. Go Dark.
Received on Tuesday, 7 April 2015 20:00:53 UTC