Re: [Simple] MSRP-ACM compatibility

["Christer Holmberg" <christer.holmberg at ericsson.com>]

Hi Ben, 

>>In fact, chapter 4.1.2 talks about TLS, and already says that RFC
>>4572 (which defines the fingerprint attribute) shall be used.
>>
>>But, I guess we can add some additional text which describes what we 
>>have been discussing.
>>
> 
>To make this interop with MSRP relays, we would need more 
>work. Relays are not involved in the SIP signaling, so there's no
opportunity for  
>them to send a fingerprint. We would need some way for endpoints to  
>get fingerprints from the relays, and include them in the signaling.

Just for my clarification: how is that related to routing based on
c/m/a=path, and possibly having a B2BUA which may modify the address
information of the ACM client's c/m/a=path?

Regards,

Christer











>There's also the fact that any fingerprint mechanism will require at  
>least integrity protection of the fingerprint, which is 
>showing itself to be controversial in the RFC4479 discussions in the
other 
>work groups.
> 
>I think that to have a useful discussion about this, we will need a  
>proposal covering this end-to-end. That doesn't necessarily mean it  
>all ends up living in the final ACM draft, but we at least need to  
>understand if we have a workable architecture.






> 
> 
> > Regards,
> >
> > Christer
> >
> > -----Original Message-----
> > From: simple-bounces at ietf.org [mailto:simple-bounces at ietf.org] On  
> > Behalf Of Christer Holmberg
> > Sent: Wednesday, April 29, 2009 3:55 PM
> > To: Hisham Khartabil; Simple WG
> > Subject: Re: [Simple] MSRP-ACM compatibility
> >
> >
> > Hi,
> >
> >> 1) We seem to have consensus that it must be possible for MSRP  
> >> devices
> >> using ACM to talk to devices using relays.
> >>
> >> We don't have consensus on how far we are willing to update
> >> RFC4975/4976 to accomplish this. We don't have consensus around any
> >> currently proposed mechanism to accomplish the requirement.
> >
> > As far as I know, the only proposed mechanism is to change the  
> > session mapping procedure, to only take the user part into account.
> >
> > But, that proposal is related to the TLS issue, so we we need to  
> > move that forward before we can make a decission.
> >
> > Regards,
> >
> > Christer
> >
> >
> >
> >
> >> 2009/4/1 Hisham Khartabil <hisham.khartabil at gmail.com>:
> >>> The MSRP-ACM draft has an open requirements question on backwards
> >>> compatibility with endpoints that use MSRP relays. In San 
> Francisco
> >>> some 20 or so people raised their hands one way or another on this
> >>> question, yet only a much smaller number have engaged in the list
> >>> discussion.
> >>>
> >>> Please offer your opinions on the following questions. We'd
> >> appreciate
> >>> the reasoning behind your opinions, rather than just yes or
> >> no votes.
> >>> We'd really like to hear from every person who raised a 
> hand in San
> >>> Francisco. If you stated an opinion at the microphone,
> >> please restate
> >>> it here.
> >>>
> >>> 1) Is there a requirement for an endpoint that uses the C-line
> >>> addressing mechanism in the MSRP-ACM draft to be able to 
> talk to an
> >>> endpoint that uses an MSRP relay (RFC 4976).
> >>>
> >>> 2) If you said yes to 1), is it acceptable to update RFC
> >> 4975 and/or
> >>> RFC 4976 to achieve compatibility?  If so, how extensively?
> >>>
> >>> Much appreciated,
> >>> Hisham
> >>>
> >> _______________________________________________
> >> Simple mailing list
> >> Simple at ietf.org
> >> https://www.ietf.org/mailman/listinfo/simple
> >>
> > _______________________________________________
> > Simple mailing list
> > Simple at ietf.org
> > https://www.ietf.org/mailman/listinfo/simple
> > _______________________________________________
> > Simple mailing list
> > Simple at ietf.org
> > https://www.ietf.org/mailman/listinfo/simple
> 
>