Re: [GnomeMeeting-list] Major ILS change

[Damien Sandras <damien sandras it-optics com>]

> My guess is that 90% (at least) of the people who are registering with
> private IP addreses are NOT behind H.323 aware routers - they are just
> users who have no idea about NAT. But I see your point.
> 
> How about this? If the signalling address in the ILS request and the
> address from which ILS request originates are both the same, or are BOTH
> public, then allow the specified port to be registered in for the ILS.
> 
> But, if the signalling address in the ILS request and the address from
> which ILS request originates are NOT the same, and the call signalling
> address is PRIVATE, then do the port check to the port specified in the
> ILS check. This will (probably) catch out users who are behind a
> firewall but do not have a H.323 aware endpoint.
> 
> I think that forcing ILS users to receive calls on port 1720 is a bad
> idea, as this will prevent more than one person behind the same NAT
> firewall from logging to the ILS. I regularly use different ports to
> allow running multiple H.323 endpoints on the same machine, and it is
> not necessary to use 1720 if you are registered with a GK.
> 

That won't happen as long as at least one of the endpoints is using
1720. But I don't understand what your approach brings as advantage to
the current one. Perhaps I'm missing something in the explanation. It
would however decrease the load of the server.

-- 
 _      Damien Sandras
(o-     
//\     It-Optics s.a.
v_/_    GnomeMeeting: http://www.gnomemeeting.org/
        FOSDEM 2004:  http://www.fosdem.org
        H.323 phone:  callto:ils.seconix.com/dsandras seconix com