Hi,
Thanks for your reply. I have a further question:
There might be applications, which do not handle registrations; the new
draft refers these applications as well, see chapter 9.3: "Note that
linking the lifetime of SIP/SigComp compartments to registration state
limits the applicability of this specification. In particular, SIP user
agents that do not register but, for example, only handle PUBLISH or
SUBSCRIBE/NOTIFY transactions are not able create SIP/SigComp
compartments following this specification."
How will these applications maintain SigComp compartments? I think the
draft should deal with this issue.
Thanks,
Dikla.
-----Original Message-----
From: Gonzalo Camarillo [mailto:Gonzalo.Camarillo at ericsson.com]
Sent: Wednesday, August 22, 2007 2:22 PM
To: Dikla Dror
Cc: rohc at ietf.org
Subject: Re: [rohc] Query about draft-ietf-rohc-sigcomp-sip-07
Hi,
previous versions of the draft addressed this type of on-the-fly
compartment creation. However, the agreement was to go for a simpler
solution that requires registrations. Therefore, the compartments would
have been created at registration time, not when A and B start
exchanging INVITEs.
Cheers,
Gonzalo
Dikla Dror wrote:
> Hi,
>
> I've sent a question concerning chapter 9 in
> draft-ietf-rohc-sigcomp-sip-07 few weeks ago, but got no answer yet.
>
> I would appreciate any feedback to the following issue (I've turned my
> question into a simpler scenario):
>
>
>
> According to chapter 9, I might encounter imbalanced compartments at 2
> different communicating endpoints in the following scenario:
>
>
>
> This scenario includes 2 endpoints (A, B), communicating over TCP.
>
>
>
>
> A B
>
> | (1)
> INVITE |
>
> Create Compartment (A1)
> |--------------------------------------------------------->| Create
> Compartment (B1)
>
>
> | |
>
> | (2)
200
> OK |
>
>
> |<---------------------------------------------------------| Use
> Compartment (B1)
>
>
> | |
>
> | (3)
> INVITE |
>
> Create Compartment (A2)
> |<---------------------------------------------------------| Use
> Compartment (B1)
>
>
> | |
>
>
> | |
>
> And so
on...
>
>
>
> As you can see endpoint A is sending an INVITE (1) toward B,
specifying
> its local urn (A's urn). However A doesn't know yet B's urn, thus it
> creates a local compartment (A1), related to the INVITE outgoing
> connection.
>
> Now B:
>
> - Responds the first INVITE with 200 OK (2), related to
> compartment (B1)
>
> - Sends an outgoing INVITE (3) to A, related to compartment
> (B1). This (3) INVITE includes B's urn.
>
>
>
> However, upon receiving (3) INVITE, endpoint A is not aware of the
> relation between B's urn and the first INVITE (1). Thus, A will
maintain
> 2 different compartments - (A1) and (A2) for these 2 sessions while B
> will maintain only one compartment (B1). This will lead to
inconsistency
> within the SigComp mechanism.
>
>
>
> How can I solve this asymmetry according to the draft?
>
>
>
> Thanks,
>
> Dikla.
>
>
>
------------------------------------------------------------------------
>
> _______________________________________________
> Rohc mailing list
> Rohc at ietf.org
> https://www1.ietf.org/mailman/listinfo/rohc