MEXT charter

Hi Behcet,

On 2007/09/28, at 0:28, Behcet Sarikaya wrote:

> Hi all,
>   Have the people in this ML paid attention to a recent
discussion  
> in netlmm on the problems related to multihomed MNs and
point to  
> point links?

The discussion of multihomed MN in NETLMM is slightly
different from  
the problem discussed in Monami6.
The netlmm issue is more general issue:
-whether an IPv6 host can have the same IP address on
multiple  
interfaces.
-how MAG/LMA can prevent assigning the same prefix to
multiple  
interfaces.

> My feeling is that in monami6 the fundamental
assumption is shared  
> link. However this may not be true even if MN attaches
to the links  
> that are traditionally shared because HA has to assign
per MN  
> prefix to MN

Why HA has to assign per MN prefix??

> and also all cellular links are point to point.

Do you mean all cellular links are managed with PMIP?

regards,
ryuji


>   Any thoughts?
>
> Behcet
>
> ----- Original Message ----
> From: RYUJI WAKIKAWA <ryuji.wakikawagmail.com>
> To: Monami6 WG <monami6ietf.org>
> Sent: Monday, September 24, 2007 1:48:12 PM
> Subject: Re: MEXT charter [was Re: [Monami6] consensus
call on  
> draft-soliman-monami6-flow-binding-04.txt]
>
> Hi Nicolas,
>
> On 2007/09/25, at 0:06, Nicolas Montavont wrote:
>
> > Hi Ryuji,
> >
> > [snip]
> >
> > On Sep 15, 2007, at 7:23 AM, Hesham Soliman
wrote:
> >
> >>>
> >>> A general question (not specific to this
document)
> >>> Monami6 WG needs to coordinate activity. I
think we
> >>> reach the consensus what we need in the WG
such as..
> >>> - registering multiple CoAs (mcoa WG
document)
> >>> - MN initiated filters registration
(draft-hesham)
> >>> - HA initiated filters registration
(supporting MH signaling
> >>> in draft-
> >>> hesham?)
> >
> > In the MEXT charter, it is written :
> >
> > A "Flow/binding policies exchange"
solution for an exchange of
> > policies from the mobile host/router to the Home
Agent and from the
> > Home Agent to the mobile host/router influencing
the choice of the
> > Care-of Address and Home Agent address. The
solution involves two
> > specifications, one for the policy format and
another for its
> > transport [both Standard Track].
>
> If this is the fixed charter, the Hesham's draft does
not satisfy the
> charter, doest it?
> It fails to support the function from the Home Agent to
the mobile
> host/router...
>
>
> > And I proposed at the last IETF to change it
because it is not
> > clear to me.
> > Here is the text I wrote on this issue:
> >
> > I think this goal is not very clear and would need
to be clarified.
> > I can read here three different goals, which are:
> >    (1) Exchange policies to influence the choice
of the CoA
> >    (2) Exchange policies to influence the choice
of the HA address
> >    (3) Specify a generic format to describe the
policies
> >
> > I think that (1) should be re-written to
"Policies enforcement from
> > the mobile host/router to the HA to influence the
choice of the
> > CoA." I suggest that because the mobile
host/router is the only one
> > to know the addresses that are mapped to its
interfaces. What we
> > need, is a protocol to indicate to the HA which
CoA to use at the
> > HA for a specific flow. The policies set up (e.g.,
I want VoIP over
> > your Wifi interface) was always out of scope of
the work in monami6.
>
> Agree with you that Policies setup is out of scope in
monami6.
> However, I cannot agree that the client is the only one
to know the
> mapping of CoA and its interface.
> If operators has wifi and wimax networks, they can
surely know which
> address is mapped to which interface.
> This is just an example, but we should not exclude this
case. (At
> least, Tasaka and koshiro expressed their support on
this approach.)
>
> > (2) We never discuss this point in the WG. Shall
we?
>
> no..
>
> regards
> ryuji
>
> >
> > (3) Can't we add an additional item to make clear
that we will have
> > 2 documents?
> >
> > Regards,
> >
> > Nicolas
> >
> >
> >>
> >> => I don't recall discussing this third
point in the meeting or
> >> getting any
> >> concensus on it. Not sure which discussion
you're referring to. I
> >> don't
> >> think draft-soliman is the right place for it,
as indicated
> >> earlier on the
> >> list.
> >>
> >> Hesham
> >>
> >>
> >>
> >>
> >>
_______________________________________________
> >> Monami6 mailing list
> >> Monami6ietf.org
> >> https:
//www1.ietf.org/mailman/listinfo/monami6
> >
> >
> > _______________________________________________
> > Monami6 mailing list
> > Monami6ietf.org
> > https:
//www1.ietf.org/mailman/listinfo/monami6
>
>
> _______________________________________________
> Monami6 mailing list
> Monami6ietf.org
> https:
//www1.ietf.org/mailman/listinfo/monami6
>
> _______________________________________________
> Monami6 mailing list
> Monami6ietf.org
> https:
//www1.ietf.org/mailman/listinfo/monami6


_______________________________________________
Monami6 mailing list
Monami6ietf.org
https:
//www1.ietf.org/mailman/listinfo/monami6

Hello Bechet,

On Thu, 2007-09-27 at 08:28 -0700, Behcet Sarikaya wrote:
> Hi all,
>   Have the people in this ML paid attention to a recent
discussion in
> netlmm on the problems related to multihomed MNs and
point to point
> links? My feeling is that in monami6 the fundamental
assumption is
> shared link. However this may not be true even if MN
attaches to the
> links that are traditionally shared because HA has to
assign per MN
> prefix to MN and also all cellular links are point to
point.
>   Any thoughts?
> 

I don't think Monami6 has the fundamental assumption of
shared link.  I
think Monami6 made no assumption whether is it shared link
or dedicated
ppp link (that may lead to the perceived, and mistaken, if I
may add,
assumption of shared link, since there are a lot more issues
one has to
consider for shared link when one is designing a protocol to
work for
both shared and point-to-point links).

/rgds
/cwng



_______________________________________________
Monami6 mailing list
Monami6ietf.org
https:
//www1.ietf.org/mailman/listinfo/monami6