Re: FW: I-D Action:draft-kaplan-sip-session-id-00.txt

> -----Original Message-----
> From: Dale Worley [mailto:[email protected]]
> Sent: Sunday, December 07, 2008 11:13 AM
>
> On Fri, 2008-12-05 at 13:43 -0500, Hadriel Kaplan wrote:
> >
> > No - read the (latest) draft.  The to-tag is in fact also used, to
> > dis-ambiguate these things.  And I note that this would mean it can't
> > be changed.  I'm on the fence whether to create a surrogate for that
> > as well. (ugh)
>
> But B2BUAs now change to-tags, so we don't have the "works in most
> current deployments" condition.  Also, to-tags may have the same problem
> as call-ids, namely that they reveal the manufacturer of the originating
> UA.  I think the mechanism needs to be revised to deal with this.

Yeah, I've been on the fence about that as I said.  I loathe adding complexity, but I admit it is probably justified for tags.

I wonder if we really need the from-tag at all for this type of thing.  Forget for the moment about any of this and just think of the current Call-ID and tags, and assume there were no B2BUA's.  Is there ever a case where a UAC would generate a request with the same Call-ID as a previous request, but a different from-tag?  The to-tag we needed to identify forked responders generating unique early-dialogs with a common Call-ID, but why did we need the UAC to generate a from-tag? (I'm feeling dumb that I can't figure out the answer)

-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