[asterisk-dev] RTP interop with Sonus: hack

[Kristian Kielhofner]

On Wed, Jan 7, 2009 at 5:52 PM, Kristian Kielhofner
<kristian.kielhofner at gmail.com> wrote:
> On Mon, Jan 5, 2009 at 2:42 PM, Joshua Colp <jcolp at digium.com> wrote:
>>
>> A few weeks ago I actually changed the RFC2833 code to do as mentioned and make it use a unique timestamp. In testing this seemed to solve the issue with a few providers (who could have been using Sonus equipment) and cause no issues with any other devices or providers. This should be in the next release of 1.4 and others so this thread may be a noop... would be nifty to test to confirm.
>>
>
> Hmmm...  After testing with Sonus gear at Level(3) and XO DTMF is
> still broken.  I have multiple packet captures for each.  Everything
> looks good to me and everyone else that has looked at them (still
> waiting on XO, actually) agrees.  Timestamps are in deed unique and
> nothing else looks broken.
>
> pcaps here:
>
> http://admin.star2star.com/xo-dtmf.pcap (XO)
> http://admin.star2star.com/dtmf5.pcap (Level(3))
>
> Any ideas?

Replying to myself...  I think I got it.  In getting all caught up on
timestamps and sequence numbers I forgot to check the SSRC.  Sure
enough, as of Josh's latest fixes RFC2833 events now have a different
SSRC than the voice RTP packets.  Not sure if this was intended but
that could very well be the "new" problem.

I'll keep looking...

-- 
Kristian Kielhofner
http://blog.krisk.org
http://www.submityoursip.com
http://www.astlinux.org
http://www.star2star.com