Re: draft-ietf-sip-199-02: comments and questions
Paul Kyzivat <[email protected]>
| Newsgroups | gmane.ietf.sip |
|---|---|
| Message-ID | <[email protected]> |
Bob Penfield wrote: > >> -----Original Message----- >> From: [email protected] [mailto:[email protected]] On Behalf Of Paul Kyzivat >> Sent: Monday, November 24, 2008 9:33 AM >> To: Christer Holmberg >> Cc: [email protected]; Brett Tate >> Subject: Re: [Sip] draft-ietf-sip-199-02: comments and questions >> >> >> >> Christer Holmberg wrote: >>> Two alternative solutions I can think of: >>> >>> 1. We mandate a forking proxy which supports 199 to store the C/R-R >>> information received from the UAC, in order to insert it in any 199 it >>> generates for that session. >>> >>> 2. We say that IF the forking proxy stores the C/R-R information >>> received from the UAC, it shall insert it in any 199 it generates for >>> that session. >> In effect the proxy is sending the response in lieu of the final >> response it is not ready to forward. IMO it ought to include whatever >> Contact and R-R is present in that final response. There isn't any need >> to *store* that information, since the final response is at hand when >> the 199 is being generated. > > The problem with this is that the non-2xx final response will not necessarily have Record-Route or Contact headers since 3261 does not require it. In fact, a strict reading of Table 2 & 3 forbids them in most error responses. If the proxy is going to generate a 199 response, it should include the same Record-Route & Contact as the original early dialog creating 1xx response. Good point. Unfortunately, that significantly raises the bar for the proxy to implement this. It needs to store a lot more state than it otherwise would need. Not a problem for a B2BUA of course, since it already has to store that. This then leaves us in a situation where a proxy incurs this significant extra cost to support 199, unless we make the inclusion of the R-R and Contact optional in the 199. I guess the Contact is already optional (though we have had discussions about that), but isn't the R-R required in all the provisional responses? Thanks, Paul _______________________________________________ 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