[asterisk-dev] RTCP RTPQOS on two SIP bridged channels

[John Todd]

It is probably not better to force your traffic (or media) through 
some other device in order to collect metrics.  I can't see in any 
circumstance where that would be 'better', 'easier', or would have 
any significant value.

There is a misnomer in the original message which may be causing 
confusion here: "SIP traffic" isn't what you're pushing through the 
"proxy".  In order to observe RTCP messages, you would have to push 
all of your RTP traffic through some other proxy and extract the 
real-time messages.  SIP is signalling, RTP is media (including 
RTCP.)  It seems that Asterisk would be a more efficient place to do 
this if you were already looking at the media as a B2BUA, and if 
Asterisk was NOT a B2BUA, then you're not going to get the 
information unless it's handed back as part of a SIP BYE  (as some 
Cisco devices do.)

If the original commenter REALLY meant SIP, then I assume this draft 
(vq-rtcpxr) has been implemented on equipment that I've never heard 
of:

   http://www.tools.ietf.org/html/draft-johnston-sipping-rtcp-summary-08


JT




At 11:59 AM -0400 2007/5/30, Alex Balashov wrote:
>On Tue, 29 May 2007, Andrew Kohlsmith wrote:
>
>>Alternatively force all SIP traffic through a proxy that does this.
>
>   That would almost certainly be easier and more portable.
>
>--
>Alex Balashov
>Evariste Systems
>Web    : http://www.evaristesys.com/
>Tel    : +1-678-954-0670
>Direct + +1-678-954-0671