many-to-many relationship in H.248

[Christian Groves]

Hello Sasha,

As Albrecht says typically there's only one control association between 
an MGC-MG pair. An MG can be split into several Virtual MGs which in 
effect allow one physical MG to have a relationship with multiple MGCs.

What scenario / functionality are you looking at? There may be another 
way via H.248 to achieve want you want.

Regards, Christian

Albrecht.Schwarz at alcatel.de wrote:

>Sasha,
>
>guess this is impossible due to the Control Association concept in H.248.
>There's only a single Control Association between a MGC-MG pair.
>Regards,
>
>Albrecht
>
>
>PS
>You should Cc "megaco at ietf.org" for H.248 specific questions.
>
>
>
>                                                                                                                                      
>                      "Sasha Ruditsky"                                                                                                
>                      <sasha at RADVISION         To:      <itu-sg16 at external.cisco.com>                                                 
>                      .COM>                    cc:                                                                                    
>                                               Subject: many-to-many relationship in H.248                                            
>                      25.01.2006 21:22                                                                                                
>                                                                                                                                      
>
>
>
>
>Hi
>
>
>In H.248 single MGC normally works with several MGs. However a particular
>MG works with a single MGC in any given moment of time.
>
>
>In a particular setup with which we are working it is necessary that an MGC
>instructs an MG to give notifications to a different MGC.
>
>
>Does somebody know a way to achieve this in current H.248?
>
>
>Do you think such way should be defined?
>
>
>Thanks,
>Sasha
>
>
>  
>