Re: draft-ietf-v6ops-addr-select-ps-01.txt and draft-ietf-v6ops-addr-select-req-02.txt WGLC

["haofeng Zhang" <hfzhang.cn@gmail.com>]

Brian,

 

Sure, this problem exists for both IPv4/IPv6 protocols. Well, it's just... I found it in IPv6 environment firstly. :)

 

I used to conduct an SIP call experiment in IPv6 network, with a SIP UA under Linux. What trouble me is that the UA gave different IPv6 addresses in SDP message for different calls. Sometimes the link-local address was included in SDP message!

 

Maybe it's bad SIP implementation. But the nature of multiple allowable IPv6 addresses per interface can cause confusion in IPv6 SIP implementation. Again, it falls into the scope of draft ICE ( http://tools.ietf.org/html/draft-ietf-mmusic-ice-16), propsed by Jean.

 

2007/7/3, Brian E Carpenter <brian.e.carpenter@gmail.com>:

Haofeng,

Why is this problem specific to IPv6? I currently have two
IPv4 addresses and can run a SIP phone on my machine. I'm sure
the same optimization problem can arise.

   Brian

On 2007-07-03 10:25, haofeng Zhang wrote:
> Should we consider another problem relating to IPv6 address selection in
> this draft?
>
> Think about this scenario in IPv6 network when SIP protocol is used:
>                                   _______________
>                                  |  IPv6 network|________ SIP Server
>                                  |______________|
>                                    /           \
>                                   /             \
>                                  /               \
>                              Router 1        Router 2
>                                 |                 |
>                                 |                 |
>                           SIP Phone tiger      SIP Phone  Deer
>                                 |                 |
>                                 |_________________|
>                                          |
>                                       Router 3
>
> In this situation, SIP Phone Tiger and Deer both have multiple addresses,
> E.g for tiger, both R1 and R3 will assign address to it, and so as sip
> phone
> Deer.
>
> Considering a sip request message initiated by tiger to sip server,
> according to RFC3484, most likely the IPv6 address assigned by Router 1
> will
> be used as the source, since the destination is sip server.  And it's
> perfectly fine  for all the signaling message between  tiger/sip server,
> and
> deer/sip server pairs.
>
> But it's not the case for the media traffic. Obviously, for both tiger and
> deer, the addresses assigned by R3 should be used for media traffic since
> they belong to the same subnet. So the problem is, should tiger's SDP
> message carry all the IPv6 addresses it has to deer so that deer can make
> the right decision; or, should a mechanism be used between tiger and
> deer to
> exchange the address selection information for SDP message?
>
> RFC3266(IPv6 support of SDP) doesn't address this. So I am not sure whether
> this problem should be addressed in this draft or another. Or have we
> already had a solution?
>
> 2007/6/21, Fred Baker < fred@cisco.com>:
>>
>> This is to initiate a two week working group last call of draft-ietf-
>> v6ops-addr-select-ps-01.txt and draft-ietf-v6ops-addr-select-
>> req-02.txt. Please read these now. If you find nits (spelling errors,
>> minor suggested wording changes, etc), comment to the authors; if you
>> find greater issues, such as disagreeing with a statement or finding
>> additional issues that need to be addressed, please post your
>> comments to the list.
>>
>> We are looking specifically for comments on the importance of the
>> documents as well as its content. If you have read the documents and
>> believe them to be of operational utility, that is also an important
>> comment to make. If you have read them and think they miss an
>> important consideration, that is very important.
>> <?xml version=" 1.0" encoding="UTF-8"?>
>>
>>
>