On Mon, Jun 15, 2009 at 3:54 PM, Robert Joly<[email protected]> wrote:
>
>
>> -----Original Message-----
>> From: M. Ranganathan [mailto:[email protected]]
>> Sent: Monday, June 15, 2009 3:49 PM
>> To: Joly, Robert (CAR:9D30)
>> Cc: sipx-dev
>> Subject: Re: [sipX-dev] Relay free sipxbridge operation for federation
>>
>> On Mon, Jun 15, 2009 at 3:31 PM, Robert Joly<[email protected]> wrote:
>> >> Hello all!
>> >>
>> >> There was a thread some time ago about relay free sipxbridge
>> >> operation and recently there has been a thread on sipx
>> federation. I
>> >> wonder what you feel about the following.
>> >>
>> >> Say you wanted to tie system A and System B together.  System A is
>> >> the "server" in this case and is providing NAT compensation to its
>> >> remote workers. System B can be configured using
>> sipxbridge to look
>> >> like a remote worker to system A.
>> >>
>> >> Clearly we don't need two media relays between the systems.
>> >> If we can support an option to suppress relaying of media, then
>> >> system B looks like a remote worker to system A.
>> >>
>> >> When a request originates from System B bound for system A via
>> >> sipxbridge, its contact and Via would get re-written but
>> sipxbridge
>> >> would leave the SDP alone.
>> >> sipxbridge on system B can  forward REFER after appropriate
>> >> re-writing of the request as it is aware that is talking
>> to another
>> >> sipx system that is capable of handling that REFER.
>> >>
>> >> It seems to me that this should be relatively easy to support in
>> >> sipxbridge. Will this scheme work in principle, as a possible
>> >> solution for federating systems? If not, what is wrong with this?
>> >
>> > I'm a bit confused by the description and I'm not sure why
>> you would
>> > want to use sipXbridge to connect two sipXecs systems in the first
>> > place?  We recommend to do this with site-to-site dialplans and the
>> > existing remote-worker solution has mechanisms in place to
>> avoid doing
>> > the double-relaying. What real-world problem are we trying solve?
>>
>> It is quite possible that a mechanism already exists and that
>> I just do not understand it so please help me understand :
>>
>> When a remote worker phone registered at site A wants to do a
>> call to a phone registered as a remote worker on  Site B,
>> could you please tell me how it would proceed ( what would be
>> the call flow and how do you avoid relaying media twice?).
>>
>>
>>
>> Thanks in advance.
>>
>
> The mechanism is described in section 4.5 of
> http://code.sipfoundry.org/browse/~raw,r=14975/sipXecs/main/sipXproxy/do
> c/NAT-traversal/NAT_Traversal_Design_Doc.doc.  It does not have a call
> flow, but I think that the mechanism can be understood from the test
> alone.


OK I think I got it. Its the x-sipx-handled.


Thanks


>



-- 
M. Ranganathan
_______________________________________________
sipx-dev mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to