> -----Original Message-----
> From: Adam Roach [mailto:[EMAIL PROTECTED]
> Sent: Monday, November 17, 2008 10:24 AM
>
> On 11/17/08 9:17 AM, Dean Willis wrote:
> > What IS important is that we understand that no such feature can be
> > made to work with an SBC that is actively disrupting it.
>
> And that's my point. Hadriel stood up and effectively said "SBCs break
> everything. If you want identity to work, you need to find something
> that SBCs don't break."

Actually, I'm pretty sure I said *B2BUA's* change the Call-ID and tags, and 
that there were far more than just SBCs.  Honestly, most every box I see traces 
from appears to be a b2bua of some form or other these days. (of course that's 
probably because of the types of networks my product's in to begin with, so 
it's a myopic view I know)  Some of these b2bua's change the call-id+tags, some 
don't.


> My understanding of set theory is somewhat rudimentary, but I am fairly
> certain that the inverse of the universe is the null set, and that the
> intersection of the null set and anything else will always be the null
> set.
> So, unless you can point out the flaw in that logic, then the only
> progress we can make is by ignoring SBCs and hope they catch up.

The set of the universe is not germane to the issue at hand.  Many B2BUA's 
change call-id+tags; that says nothing about whether they change other things.  
And it says nothing about their motivation for the changes.

SBC's cannot "catch up" to not change call-id's - it's what their owners want 
them to do.  Speaking as just one vendor, we don't change call-id's and tags by 
*default*.  Virtually EVERYONE changes that default behavior and configures us 
to change the call-id and tags as soon as they install the system.  That 
doesn't mean they will change anything no matter what - they have specific 
motivation for changing the call-id.  The IETF SIP WG gave them that 
motivation, by putting an IP Address in it.  That was a mistake.  It was a big 
mistake.

-hadriel
_______________________________________________
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