[Freeswitch-users] RTP Timestamp changes after REFER - Question on RFC conformity

[Christian Benke]

Hi!

Here's a trace: http://poab.org/misc/one_way_audio_internal_g711.pcap
(or http://filedump.org/files/yMnIRQUC1345735448.html for the archive)

The REFER happens at Paket 1305, Time 25.662729

The RTP-Paket that are sent to the C-endpoint after the REFER are
continuous from A's point of view(As it is the continuation of the
A-B-stream), but not from C's point of view.

Regards,
Christian

On 23 August 2012 17:18, Anthony Minessale <anthony.minessale at gmail.com> wrote:
> Do you have a pcap of it not setting the mark bit when the timestamp
> changes because the code will always set the mark bit when the packet
> is about to send is not exactly the next packet it expected to send
> after the previous one/
>
>
> On Thu, Aug 23, 2012 at 10:04 AM, Christian Benke <benkokakao at gmail.com> wrote:
>> Hi!
>>
>> I had a problem with VOP Softclient(www.voiceoperatorpanel.com)  where
>> the callee had one-way-audio after a call was REFERred to him. After
>> lots of tracing, debugging and hairpulling i realized the problem lies
>> in the changing rtp-timestamp when the RTP-stream is switched from the
>> middleman to the initial call after the REFER. The Softclient was not
>> able to cope with the changed timestamp and ignored the incoming
>> RTP-packets, leading to no audio for the callee.
>>
>> I was eventually able to solve the problem by activating
>> rtp-rewrite-timestamps on the profile(Also added it to
>> http://wiki.freeswitch.org/wiki/RTP_Issues#Voiceoperatorpanel_VOP).
>>
>> However, i would like to know if FreeSWITCH/Sofia is working according
>> to the RFC and if the Softclient is to blame for the problem(So i can
>> file a bugreport with them).
>>
>> In this thread, Brian West states that  it's ok to skip forward in
>> timestamps as long as the marker-bit is set:
>> http://lists.freeswitch.org/pipermail/freeswitch-users/2010-July/060333.html
>> However, the Marker-bit is not set by FreeSWITCH when the REFER occurs.
>>
>> I didn't find this stated in http://tools.ietf.org/html/rfc3550(But
>> "timestamp" is mentioned a lot, so i may have missed it) but there's
>> this bug-report for Asterisk, where the exact same problem is
>> described and eventually handled:
>> https://issues.asterisk.org/view.php?id=17007
>>
>> Could someone with more insight please elaborate?
>>
>> Best regards,
>> Christian
>>
>> _________________________________________________________________________
>> Professional FreeSWITCH Consulting Services:
>> consulting at freeswitch.org
>> http://www.freeswitchsolutions.com
>>
>> 
>> 
>>
>> Official FreeSWITCH Sites
>> http://www.freeswitch.org
>> http://wiki.freeswitch.org
>> http://www.cluecon.com
>>
>> FreeSWITCH-users mailing list
>> FreeSWITCH-users at lists.freeswitch.org
>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
>> http://www.freeswitch.org
>
>
>
> --
> Anthony Minessale II
>
> FreeSWITCH http://www.freeswitch.org/
> ClueCon http://www.cluecon.com/
> Twitter: http://twitter.com/FreeSWITCH_wire
>
> AIM: anthm
> MSN:anthony_minessale at hotmail.com
> GTALK/JABBER/PAYPAL:anthony.minessale at gmail.com
> IRC: irc.freenode.net #freeswitch
>
> FreeSWITCH Developer Conference
> sip:888 at conference.freeswitch.org
> googletalk:conf+888 at conference.freeswitch.org
> pstn:+19193869900
>
> _________________________________________________________________________
> Professional FreeSWITCH Consulting Services:
> consulting at freeswitch.org
> http://www.freeswitchsolutions.com
>
> 
> 
>
> Official FreeSWITCH Sites
> http://www.freeswitch.org
> http://wiki.freeswitch.org
> http://www.cluecon.com
>
> FreeSWITCH-users mailing list
> FreeSWITCH-users at lists.freeswitch.org
> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
> http://www.freeswitch.org