[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Sip] Delivering request-URI and parameters to UAS via proxy



 
Inline

Youssef

> -----Message d'origine-----
> De : Paul Kyzivat [mailto:pkyzivat at cisco.com] 
> Envoyé : mercredi 30 janvier 2008 16:12
> À : CHADLI Youssef RD-CORE-ISS
> Cc : drage at alcatel-lucent.com; sip at ietf.org
> Objet : Re: [Sip] Delivering request-URI and parameters to 
> UAS via proxy
> 
> 
> 
> youssef.chadli at orange-ftgroup.com wrote:
> > I think there is a misunderstanding on these use cases:
> 
> Quite possible.
> 
> >  - The intermediate entity in the customer side does 
> perform only one registration to the network.
> >    The AoRs associated to the served terminals inside the 
> client domain are registred automatically
> >    by the home network, using implicit registration for 
> example. In fact, it's not conceivable to
> >    register individually all the users of a corporate 
> network for example.
> 
> You seem to be describing one of the models covered by the 
> SIPForum in their SIPConnect specification. Is that what you mean?

The model described in Sipconnect is an example of such configuration. As I mentionned before there other cases such as home networks and other configurations of connecting corporate networks.
 
>  From the point of view of the registrar this seems similar 
> to IMS implicit registration, except that the scale is much 
> larger - there may be *very many* implicit registrations 
> generated by this one explicit registration.

Agreed

> 
> > - The "intermediate entity" should be able to route 
> incoming requests inside its domain based on
> >   standard RFC3261 behaviour, that is, based on the Request-URI.
> 
> I find the case much less compelling than the IMS case.
> 
> IMO this is an abuse of REGISTER to a purpose for which it is 
> not well suited.
> 
> Treating it as registration means that there is an 
> *expectation* that the R-URI will be translated. That isn't 
> really appropriate in this case. The registration doesn't 
> specify the mapping anyway - it is configuration in the 
> network of all the addresses that belong to the "pbx" that 
> does that. This is more appropriately viewed as just a 
> routing operation, and so pushing the address of the 
> enterprise server as a Route header is by far the best 
> choice. Its only because this has been done using REGISTER 
> that brings up how to preserve the old address when it is translated.

Agreed
> 
> 	Thanks,
> 	Paul
> 
> > Best regards
> > 
> > Youssef
> > 
> > 
> >   
> >  
> > 
> >> -----Message d'origine-----
> >> De : Paul Kyzivat [mailto:pkyzivat at cisco.com] Envoyé : lundi 28 
> >> janvier 2008 21:46 À : CHADLI Youssef RD-CORE-ISS Cc : 
> >> drage at alcatel-lucent.com; sip at ietf.org Objet : Re: [Sip] 
> Delivering 
> >> request-URI and parameters to UAS via proxy
> >>
> >> These cases can all be handled by having intermediate entity use a 
> >> unique user part, together with its own domain name, to 
> construct a 
> >> contact address for each terminal UA that it is registering. Then 
> >> when it receives an incoming request to one of these uris it can 
> >> simply translate it, using the user part as a key to its own local 
> >> mappings.
> >>
> >> We don't need anything new to solve these cases.
> >>
> >> 	Paul
> >>
> >> youssef.chadli at orange-ftgroup.com wrote:
> >>>  
> >>> There are other cases similar to what is described in
> >> section  2.1 Unknown Aliases of J. Rosenberg draft that should be 
> >> taken into account.
> >>> Those cases are where the customer side is composed of
> >> several SIP terminals that access the network through an 
> intermediate 
> >> entity which registers their associated aliases with its contact 
> >> address. Thus, such client side is seen from the network 
> as a single 
> >> client having several aliases (aggregated endpoints). 
> Moreover, such 
> >> client side may be a mini private network composed of several 
> >> entities. In these configurations, the intermediate entity is in 
> >> charge of routeing incoming calls inside the client domain and may 
> >> need to behave as a SIP proxy for incoming SIP messages.
> >>> As examples of such configuration:
> >>> - Corporate networks: in that case the PBX register all the
> >> served individual user identities with its contact address 
> and routes 
> >> incoming calls toward the individual called users.
> >>> - Home networks: the Home Gateway may need to register on
> >> behalf of all the served terminals.  The Home Gateway is 
> in charge of 
> >> routeing incoming calls toward the individual called users.
> >>> J. Rosenberg draft seems give a good solution to take into
> >> account these configurations.
> >>>    
> >>> Best regards,
> >>>
> >>> Youssef
> >>>
> >>>
> >>>> -----Message d'origine-----
> >>>> De : DRAGE, Keith (Keith)
> >> [mailto:drage at alcatel-lucent.com] Envoyé : 
> >>>> lundi 14 janvier 2008 17:58 À : sip at ietf.org Objet : [Sip]
> >> Delivering
> >>>> request-URI and parameters to UAS via proxy
> >>>>
> >>>> (As WG chair)
> >>>>
> >>>> In fulfilment of our charter items of
> >>>>
> >>>> Dec 2007    Delivering request-URI and parameters to UAS 
> >> via proxy to
> >>>> WGLC  
> >>>> Feb 2008    Delivering request-URI and parameters to UAS 
> >> via proxy to
> >>>> IESG (PS)
> >>>>
> >>>> We now have a couple of proposals on the table for solving the 
> >>>> problem.
> >>>>
> >>>> The original draft from Jonathan and which led to the
> >> creation of the
> >>>> charter items by the WG is unfortunately expired, but is at:
> >>>>
> >>>> 
> http://tools.ietf.org/id/draft-rosenberg-sip-ua-loose-route-01.txt
> >>>>
> >>>> The alternative document from Christer, etc is at:
> >>>>
> >>>> http://www.ietf.org/internet-drafts/draft-holmberg-sip-target-
> >>>> uri-delive
> >>>> ry-00.txt
> >>>>
> >>>> We obviously need to make a decision between the two 
> approaches so 
> >>>> please attempt to address the following specific points via the 
> >>>> mailing
> >>>> list:
> >>>>
> >>>> 1)	Problem cases: These are summarised in section 4.4 of
> >>>> draft-rosenberg-sip-ua-loose-route-01 and from my read of
> >> the other
> >>>> draft, I don't believe that this draft adds any others.
> >>>> If you believe there are other cases that should be 
> covered by the 
> >>>> solution, then please identify them. If there is support
> >> on any new
> >>>> problem cases, I would encourage the authors of both 
> drafts to add 
> >>>> text concerning these problem cases.
> >>>>
> >>>> 2)	Clarifications: If for any reason you don't 
> understand either
> >>>> draft, or believe that there are technical issues that are not 
> >>>> represented in the current draft, please post your questions / 
> >>>> comments to the list. I would encourage authors of both 
> drafts to 
> >>>> revise as frequently as appropriate to reflect the current
> >> state of
> >>>> discussion.
> >>>>
> >>>> 3)	Support for either position. If you wish to 
> indicate support for
> >>>> either position please do so, but please accompany this is
> >> technical
> >>>> reasoning as to why you have this position, as that will
> >> help other
> >>>> members of the WG form a position.
> >>>>
> >>>> I would encourage as much list discussion as possible before 
> >>>> Philadelphia. I suspect we will need to have a discussion at the 
> >>>> face-to-face meeting in Philadephia, but list discussion
> >> is essential
> >>>> prior to that. If we can solve this positions on list,
> >> then well and
> >>>> good, and even better (that is how we are meant to make 
> decisions).
> >>>>
> >>>> Regards
> >>>>
> >>>>
> >>>> Keith
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
> >>>> This list is for NEW development of the core SIP Protocol Use 
> >>>> sip-implementors at cs.columbia.edu for questions on 
> current sip Use 
> >>>> sipping at ietf.org for new developments on the application of sip
> >>>>
> >>>
> >>> _______________________________________________
> >>> Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
> >>> This list is for NEW development of the core SIP Protocol Use 
> >>> sip-implementors at cs.columbia.edu for questions on current sip Use 
> >>> sipping at ietf.org for new developments on the application of sip
> >>>
> > 
> 


_______________________________________________
Sip mailing list  https://www1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementors at cs.columbia.edu for questions on current sip
Use sipping at ietf.org for new developments on the application of sip