1174689 - Subsequent setLocalDescription fails with InvalidSessionDescriptionError

Status: RESOLVED INCOMPLETE

(Core :: WebRTC, defect)

Description

[Michael Froehlich]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_3) AppleWebKit/600.6.3 (KHTML, like Gecko) Version/8.0.6 Safari/600.6.3

Steps to reproduce:

I initiated a webrtc call from Firefox to Chrome, where FF 38.0.5 is only receiving audio. During the session I wanted to upgrade to bidirectional audio, however setLocalDescription failed with:

DOMException [InvalidSessionDescriptionError: "Invalid description, no ice-ufrag attribute" code: 0 nsresult: 0x0]



Actual results:

Firefox is making the offer in both cases. The first one with OfferToReceiveAudio: true, the second one providing the local stream into the session.

First offer SDP:

v=0
o=mozilla...THIS_IS_SDPARTA-38.0.5 1789871440271733710 0 IN IP4 0.0.0.0
s=-
t=0 0
a=fingerprint:sha-256 A2:05:F8:7D:53:60:F1:9B:71:A8:A4:F5:90:9F:17:18:A7:AF:5F:CE:0C:26:ED:8B:02:71:82:2D:4B:2F:94:3B
a=group:BUNDLE sdparta_0
a=ice-options:trickle
a=msid-semantic:WMS *
m=audio 9 RTP/SAVPF 109 9 0 8
c=IN IP4 0.0.0.0
a=recvonly
a=extmap:1 urn:ietf:params:rtp-hdrext:ssrc-audio-level
a=ice-pwd:2beebda3c141cef0ded6913d0a86caf7
a=ice-ufrag:fa037e22
a=mid:sdparta_0
a=rtcp-mux
a=rtpmap:109 opus/48000/2
a=rtpmap:9 G722/8000/1
a=rtpmap:0 PCMU/8000
a=rtpmap:8 PCMA/8000
a=setup:actpass

Chrome answer:

v=0
o=- 9126077397863599194 2 IN IP4 127.0.0.1
s=-
t=0 0
a=group:BUNDLE sdparta_0
a=msid-semantic: WMS 64527DBB-A018-4B44-B6AA6988D3199DE1
m=audio 1 RTP/SAVPF 109 0 8
c=IN IP4 0.0.0.0
a=rtcp:1 IN IP4 0.0.0.0
a=ice-ufrag:YWb9jS/zeznEu1u5
a=ice-pwd:LZSXXa9bn4HTjF3BpaLkIR71
a=fingerprint:sha-256 94:BE:F3:87:BC:30:11:C7:FD:9F:15:1A:A8:6A:A3:6C:E0:A8:3A:8B:BB:1D:62:1D:5F:26:4E:2B:85:6D:39:01
a=setup:active
a=mid:sdparta_0
a=extmap:1 urn:ietf:params:rtp-hdrext:ssrc-audio-level
a=sendonly
a=rtcp-mux
a=rtpmap:109 OPUS/48000/2
a=fmtp:109 minptime=10
a=rtpmap:0 PCMU/8000
a=rtpmap:8 PCMA/8000
a=maxptime:60
a=ssrc:2924590551 cname:Q1t0aRZxEDgXRkyA
a=ssrc:2924590551 msid:64527DBB-A018-4B44-B6AA6988D3199DE1 64527DBB-A018-4B44-B6AA6988D3199DE1
a=ssrc:2924590551 mslabel:64527DBB-A018-4B44-B6AA6988D3199DE1
a=ssrc:2924590551 label:64527DBB-A018-4B44-B6AA6988D3199DE1

Second offer SDP:

v=0
o=mozilla...THIS_IS_SDPARTA-38.0.5 1120799112872777402 1 IN IP4 0.0.0.0
s=-
t=0 0
a=fingerprint:sha-256 1E:D0:99:A7:B8:89:66:A2:51:DD:37:2C:A7:7C:35:A8:3B:18:D4:6B:12:D2:5F:61:80:DE:3A:DF:25:74:CF:4A
a=group:BUNDLE sdparta_0
a=ice-options:trickle
a=msid-semantic:WMS *
m=audio 1 RTP/SAVPF 109 9 0 8
c=IN IP4 0.0.0.0
a=candidate:3010947332 1 udp 2097152000 10.1.136.103 45000 typ host generation 0
a=candidate:4260664820 1 tcp 1476395008 10.1.136.103 45000 typ host generation 0
a=sendrecv
a=extmap:1 urn:ietf:params:rtp-hdrext:ssrc-audio-level
a=ice-pwd:ed0a086945eaf3146120bd0b5bf1e539
a=ice-ufrag:e30d6b0e
a=mid:sdparta_0
a=msid:{18257102-b9fd-e64e-8b29-865abb845028} {46d3254f-d955-1f41-9daa-6b00ffe89d61}
a=rtcp-muxa=rtpmap:109 opus/48000/2
a=rtpmap:9 G722/8000/1
a=rtpmap:0 PCMU/8000
a=rtpmap:8 PCMA/8000
a=setup:actpass
a=ssrc:1565517672 cname:{e93b630b-6886-654b-96a6-e08968adb6a5}

setLocalDescription failed with error: DOMException [InvalidSessionDescriptionError: "Invalid description, no ice-ufrag attribute" code: 0 nsresult: 0x0]

Even though I see ice-ufrag in the first and second offer.


Expected results:

Local description should have been applied without issues.

Comment 1

[Maire Reavy [:mreavy]]

Can one of you look at this when you get back from PTO?  I'm looking to confirm and triage this.

Comment 2

[Byron Campen [:bwc]]

This doesn't make any sense. That second SDP has a different fingerprint, ice-ufrag, and ice-pwd. All three are things that should stay the same in the current codebase. Can you post a test-case?

Comment 3

[Nils Ohlmeier [:drno]]

How come that the second SDP offer contains a TCP ICE candidate even though Fx 38 does not support ICE TCP (yet)?

Comment 4

[Maire Reavy [:mreavy]]

> This doesn't make any sense. That second SDP has a different fingerprint,
> ice-ufrag, and ice-pwd. All three are things that should stay the same in
> the current codebase. Can you post a test-case?

To the bug reporter -- Hi Michael, Can you post a test-case?  Else there's nothing actionable for us here, and I'll need to resolve this bug as INCOMPLETE.  Thanks.

Comment 5

[Maire Reavy [:mreavy]]

Marking this resolved incomplete because we can't move forward without more info from the bug reporter, and we've had an unanswered needinfo into him/her for over a month.