[asterisk-bugs] [JIRA] (ASTERISK-27826) No sound in some calls (webrtc)

[George Joseph (JIRA)]

    [ https://issues.asterisk.org/jira/browse/ASTERISK-27826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=243189#comment-243189 ] 

George Joseph commented on ASTERISK-27826:
------------------------------------------

OK, so a simple webrtc -> MOH has the issue?
What codecs are enabled for the webrtc client and what format are the MOH files?
Actually, if you can give me the full output of "pjsip show endpoint <webrtc_client>" that would help greatly.

Also if possible, can you get a witeshark capture of both the SIP and RTP flow?  I know it's encrypted but I want to look at the TLS/DTLS exchange.
At the same time, if you can get the output from "pjsip set logger <host>", where <host> is the webrtc client, that'd help as well.




> No sound in some calls (webrtc)
> -------------------------------
>
>                 Key: ASTERISK-27826
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-27826
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Resources/res_rtp_asterisk
>    Affects Versions: 15.3.0
>            Reporter: Mikhail Ivanov
>            Assignee: Unassigned
>              Labels: webrtc
>
> I have a problem with incoming (may be with outgoing too, not sure) calls to WebRTC clients (based on jssip.net library)
> Sometimes (2-5% of all incoming calls) I have no sound (on both sides) on incoming calls.
> RTP is going fine in both sides (local network)
> If I turn on mixMonitor on Asterisk, I can see only noise in call (looks like a problem with srtp keys, but not sure)
> https://www.dropbox.com/s/41nmwqhg0chcwl7/cf626000ac4601445d6cee3cd909188d.mp3?dl=1
> Asterisk 15.3.0, JsSIP 3.2.8, tested in Chrome, Chromium and Firefox
> If I turn off rtp encryption 
> webrtc = no 
> rtcp_mux = yes 
> use_avpf = yes 
> ice_support = yes 
> media_encryption = no
> and 
> --disable-webrtc-encryption in crome
> everything is fine, yes, it's workaround but not a solution



--
This message was sent by Atlassian JIRA
(v6.2#6252)