Any solution that will not work through conformant (:-) B2BUA's will be
useless in practice. It is the same kind of thing as ignoring NATs.

/Hans Erik

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of [EMAIL PROTECTED]
Sent: Wednesday, March 12, 2008 3:48 PM
To: [EMAIL PROTECTED]
Cc: [email protected]
Subject: Re: [BLISS] Call-completion design question: End-to-end
parameters

   From: Paul Kyzivat <[EMAIL PROTECTED]>

   > The advantage of using Call-Id to identify calls is that all calls
   > already have Call-Ids.  Without using Call-Id as the default
   > identifier, every call originator has to be revised to add another
   > identifier to every call.

   Perhaps one could make the case that if you are trying to anonymize
your 
   call then call completion *shouldn't* work. If so, then we only need 
   deal with cases where it isn't anonymized.

I think that John's case happens in non-anonymized calls as well.

   Then this would become yet another requirement on B2BUAs - that
   they not mess with callids.

The irony is that one of the few RFC 3261 restrictions on B2BUAs is that
they don't use the same Call-Id on the outgoing leg that was used on the
incoming leg.  And many B2BUAs violate that.  But the violaters do not
cause a problem for the CC design; it's the conformant B2BUAs that are
the problem.

Dale
_______________________________________________
BLISS mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/bliss
_______________________________________________
BLISS mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/bliss

Reply via email to