Dean Willis writes:

 > Why do I get the feeling that you won't be happy even when every UA is  
 > forced to register every contact it has or might obtain have with  
 > every registrar that ever existed or will exist ;-)?

dean,

that is not true.  i sincerely hoped that outbound would help me to
implement redundant proxy/registrar without a need to for load
balancers, database clusters, or forwarding of requests between
proxies simply by being able to assume that if i have two ob proxies,
each ua would register all contacts that it has via both of them.  

this is a VERY simple, easy to understand and implement requirement
compared to anything else this working group has ever produced.

-- juha



 > 
 > On Oct 6, 2008, at 11:43 AM, Juha Heinanen wrote:
 > 
 > > Dean Willis writes:
 > >
 > >> Assume a UA has two different interfaces, and it registers each of  
 > >> its
 > >> interfaces with a different proxy/registrar. Further assume that  
 > >> those
 > >> two proxy registrars collude with respect to mapping the user's AUR
 > >> onto those contacts.
 > >
 > > that "collude" thing is exactly what i tried to avoid.  if ua is
 > > required to register all of its contacts via both proxies, then
 > > redundancy of the service can be implemented by running the
 > > two proxy/registrars completely independent of each other thus greatly
 > > simplifying the implementation.
 > >
 > > if that requirement is not there, ob becomes totally useless again and
 > > there was no advantage that the text was changed according to my
 > > wishes.  you may as well go and change it back to what it was before.
 > 
_______________________________________________
Sip mailing list  https://www.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [EMAIL PROTECTED] for questions on current sip
Use [EMAIL PROTECTED] for new developments on the application of sip

Reply via email to