Re: SDP in H248 v/s Megaco

Atul,

which H.248 profile specification have you implemented?

There are profiles (like 3GPP H.248 Mc profile according 29.232) using
"tunneling mechanisms" for bearer control (e.g. -> IPBCP Q.1970 with
tunnel protocol Q.1990). 
[Background: The Bearer Control Tunneling Package defined in Annex A,
section A.7 of Q.1950 defines a signal and observed event parameter in
order to pass SDP between the MG and MGC.]

Thus, pls confirm whether you use such a tunnel or not.

-Albrecht 
 

> -----Original Message-----
> From: [email protected] 
> [mailto:[email protected]] On Behalf Of atul garg
> Sent: Dienstag, 29. Juli 2008 12:11
> To: [email protected]
> Cc: [email protected]
> Subject: [Megaco] SDP in H248 v/s Megaco
> 
> Hi,
> 
> Can someone help me with the following doubts in H.248?
> 
> 1. As per my understanding ,in H.248, when the MG1 indicates 
> to MGC (via off hook) that it wants to communicate to MG2, 
> MGC sends an ADD Request to MG1. MG1 creates a context and 
> Ephemereal TerminationID and allocates resources for the 
> same. It sends this information (about media resources i.e. 
> IP and port) to the MGC in SDP in its Local Descriptor. MGC 
> conveys this information to MG2 via SDP's Remote descriptor 
> when it sends the ADD Request to MG2. MG2, in turn, allocates 
> the resources and conveys its IP and port for communication 
> in the ADD Reply's SDP (Local Descriptor), which is then 
> passed to MG1.So, whenever SDP is used to convey the 
> connection information, it is done in the Local and Remote 
> descriptors. 
> 
> Now the question is what is the use of tunneling SDP 
> information in Signal and ObservedEvent Descriptor?
>  
> I have a large amount of network traffic, and in all the 
> buffers, SDP information NEVER comes in Media descriptor's 
> local and remote. It always comes tunneled in Signal and 
> ObservedEvent descriptors. So, can I somehow use this 
> tunneled SDP information instead of local and remote descriptors? 
> 
> 2. As per the protocol, every ADD request from MGC to MG 
> should contain two ADD commands: 
> (i)	Create a context and ADD a well-defined physical 
> termination to it. 
> (ii)	Create an Ephemereal Termination, allocate IP and port 
> and add it to the context created in (i).
>    According to this, every buffer containing ADD request 
> must have two ADD commands. However, in H.248 traffic, I have 
> observed that the request to add physical termination is not 
> present along with the request for creating an ephemereal 
> termination. Can someone explain the reason why this could be 
> happening?
> 
> 3. In H.248, NOTIFY request should be sent to the physical 
> Termination. However, in the buffers I have, NOTIFY request 
> is sent on the Ephemereal termination. 
> 
> In MEGACO's traffic, I have observed that:
> (i)	ADD Request always contains two commands: to add 
> physical termination and to create andadd epemereal termination.
> (ii)	NOTIFY is received on the physical termination.
> 
> Then, why doesn't the H.248 traffic conform to the 
> requirements mentioned in the protocol?
> 
> 
> 
>       Bring your gang together. Do your thing. Find your 
> favourite Yahoo! group at http://in.promos.yahoo.com/groups/
> _______________________________________________
> Megaco mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/megaco
>