Notes on SIP Session 1

Reported by Paul Kyzivat

SIP WG - IETF-62 – March 7, 2005
Scribe: Paul Kyzivat

o GRUU Issues
    (draft-ietf-sip-gruu-03.txt)
    Jonathan Rosenberg

It was pointed out that the UML diagram in the current version has some
incorrect cardinalities. Jonathan said he would fix. Jonathan explained
changes around support of SIPS GRUUs - it is now as good, and bad, as
3261 itself. Need to write a draft clarifying how to handle SIPS in 3261.

The new version changed the handling when a new contact is registered
with the same instance id as an already registered contact. It now
automatically removes the earlier registration. The draft points out
this provides a convenient way to unregister any contact with a
particular instance id. It was pointed out by someone that this
"feature" will probably be broken by the extension elsewhere to allow
multiple contacts with same instance id. There was suggestion to soften
the description of this feature.

Jonathan highlighted an issue of Record-Route between edge and home
proxies and looping around gruu. Concern was raised about rewriting the
R-R even though 3261 allows this in some cases. Discussed alternatives,
but didn’t resolve. That is TBD - discuss on list.

Discussed issue about getting AOR from gruu - why this is/isn’t a need.
Objections were raised to the proposed compromise in the draft. Various
other alternatives discussed. Cullen Jennings didn't see a compelling
reason for it. Paul Kyzivat said he no longer found this a need-
solutions other than GRUU were sufficient. Aki Niemi and Orit Levin
still expressed a desire for this feature. Dean, as chair, said we have
no requirement for this, and a domain can do it anyway if it wants.

Discussed relationship of identity to gruu. Jon Peterson thought a gruu
isn’t an identity. Discussion halted without resolution due to time.
Need to work this offline.

o Non invite xactions
    (draft-sparks-sip-nit-future-01.txt)
    Robert Sparks

Agreed not to pursue Alternative B.

Agreed that the Timeleft alternative was also not worth pursuing.

Success/failure caching: Robert said we didn’t have a way to tell when
something becomes available. But Jonathan Rosenberg suggested that the
Jennings outbound draft (draft-jennings-sipping-outbound-01.txt)
provides a way to do this – via sending STUN keepalives.
[ED: this is what I wrote, but it now doesn't make sense to me.]

Something about a “try again later” mechanism. Conclude the group does
want to continue to pursue a general mechanism for this. Robert Sparks
proposed using a blacklist for now. Rohan Mahy proposed replacing this
with a “push to back” mechanism, because it works even if there is only
one alternative. Rohan, as chair, asked to defer liveness testing to the
list.

o REFER Extensions
    (draft-ietf-sip-refer-with-norefersub-01.txt)
    Orit Levin

There was a question about how refer issuer knows if it can use this
feature. Various alternatives shown. Jonathan Rosenberg argued why using
Supported has wrong semantics. Sent team out of the room to work on this.

Decision from the team: keep option tag, add header field used in
request to ask for the feature and in response to acknowledge it.
Discourage use of Require of the option.

o Identity -- Request
    (draft-ietf-sip-identity-04.txt) –
    Jon Peterson

Issue raised about what proxy should do if it gets one of the new
responses. Concluded that the identity server should be able to retry
the request with correction applied. And, if there is a forking proxy
beyond an identity server, it should choose recoverable errors to
forward back in order that identithy server may recover.

Discussed the importance of signature including the display name.
Decided that signing over the display name will be done.

There had been a special case that identity is not applied to REGISTER.
Jonathan Rosenberg showed case where it was useful. So it will be made
legal.

o Identity -- Response
    (draft-peterson-message-identity-00.txt)
    Jon Peterson

Jon presented alternative ways forward. Jonathan Rosenberg said major
threat is if someone fakes a connected party id, assuming we had that,
and that such is a requirement. Rohan Mahy thought History-Info is the
right solution and said he plans to write up something about it.

o SIP Resource Priority
    (draft-ietf-sip-resource-priority-06.txt)
    James Polk

James noted that this is two weeks shy of the 5th anniversary of this
draft. Version -07 to be published next week. They hope to WGLC it.

o Accept-Disposition
    (draft-camarillo-sip-accept-diposition-00.txt)
    Gonzalo Camarillo

Lots of discussion about this. Rather than pursue this approach, for the
application (exploder) that stimulated this draft there was a decision
to use an option tag to negotiate support for the needed combination of
features.

Separately, there was interest by some (at least Gonzalo and Paul) in
starting to investigate the underspecification of Content-Disposition.