Re: : The Diameter Maintanence and Extentions WG proposed charter

All,
Sorry, I've been in email transition and have been a little late in
catching up with all the emails.

I agree with Hannes and John that we can make a decision once we have
the details clearer.

As far as mip6 is concerned, even though there are the two
bootstrapping drafts, the work on auth protocol bootstrapping is yet
to be done. So, I don't think we know all the AAA requirements just
yet. To write separate AAA documents to handle each case would be a
waste.

I'm still not fully clear on why we cannot just re-use the Diameter
NASREQ application for FMIPv6 if that is what mip6 is using. Where the
NAS is located is different in each case, but that doesn't seem to be
a major issue to me.

Perhaps, we should plan on having a discussion at the next IETF with
the relevant mobility folks and collect requirements - that would be a
good start.

Thanks,
Vidya

On 12/1/05, Tschofenig, Hannes <[email protected]> wrote:
> hi julien,
>
> to me it still seems that we do not understand all details yet and theref=
ore i am not sure whether we need a new application for either fmip or mip6=
 aaa interaction.
>
> cannot we make a decision about one or two documents once we know the det=
ails much better?
>
> ciao
> hannes
>
> > -----Urspr=FCngliche Nachricht-----
> > Von: [email protected] [mailto:[email protected]]
> > Im Auftrag von Julien Bournelle
> > Gesendet: Donnerstag, 1. Dezember 2005 11:36
> > An: Tschofenig, Hannes
> > Cc: Julien Bournelle; [email protected];
> > [email protected]; [email protected]
> > Betreff: Re: [AAA-WG]: The Diameter Maintanence and
> > Extentions WG proposed charter
> >
> > Hi hannes,
> >
> > On Thu, Dec 01, 2005 at 10:57:13AM +0100, Tschofenig, Hannes wrote:
> > > hi julien,
> > >
> > > we are not fully sure what fmip and mip really requires.
> >
> >  for mip6 bootstrapping, we have 2 solutions: integrated and split.
> >
> >  In
> >  the integrated scenario, we assume that the MN is authenticated for
> >  network access and thus we implicitely assume that Diameter
> > NASREQ/EAP
> >  is used. At the end of the authentication, the AAAH sends a
> > HA address
> >  to the NAS. So, a priori we only need this AVP.
> >
> >  In the split scenario, the AAA exchange occurs between the
> > HA and AAA.
> >  And IKEv2 is used between MN and HA. So Diameter EAP should be used
> >  between HA and the AAA to carry the EAP packets. As it is a
> > HA and not
> >  a NAS, we "may" define a specific application.
> >
> >  for fmip6, we might reuse Diameter NASREQ, but why not define a new
> >  Diameter Application since what we want to do is really do
> > AAA for the
> >  FMIP6 service and not for network access.
> >
> > > we have started with some thoughts but more input is needed.
> >
> >  I agree
> > >
> > > i do not see a problem with the fact that the protocols are
> > developed in
> > > different working groups.
> >
> >  the only problem that I can see is time. Maybe of one these
> > 2 WGs might
> >  require AAA before the other. If we have a single document,
> > we'll have
> >  to wait for both agreements. that's just my concern.
> >
>
>
> ciao
> hannes
>
> >  Julien
> >
> > >
> > > ciao
> > > hannes
> > >
> > > > On Thu, Dec 01, 2005 at 11:07:21AM +0200,
> > > > [email protected] wrote:
> > > > > Julien,
> > > > >
> > > > > >> For the purposes of the DiME WG proposal, I think I am
> > > > still lumping
> > > > > >> MIPv6, FMIPv6 & MIPv6 bootstrapping into one topic.  I
> > > > think having a
> > > > >
> > > > > >> single AAA document that addresses these points probably
> > > > would be
> > > > > >> useful.
> > > > > >
> > > > > > I don't think that we can have one document for all these
> > > > > > things. You  can use Mobile IPv6 without FMIP6.
> > > > >
> > > > > I look at FMIPv6 as an optional extension for MIPv6.  So,
> > > > if you need
> > > > > AAA support for FMIPv6, then you probably will need to
> > also have AAA
> > > > > support for MIPv6.  However, if you are looking at only
> > MIPv6, then
> > > > > any extensions to support FMIPv6 would be purely optional.
> > > >
> > > >  yes you're right.
> > > >
> > > >   On the other hand, what makes me think that 2 documents could be
> > > >   better are the following arguments:
> > > >
> > > >   - FMIP6 is from the mipshop WG whereas MIP6 is from the
> > mip6 WG.
> > > >   - FMIP6 probably requires a new Diameter Application
> > whereas MIP6
> > > >     bootstrapping requires only AVPs.
> > > >
> > > >  Julien
> > > >
> > > > >
> > > > > >> However, as the MIP/FMIP work is still on-going, we can
> > > > address this
> > > > > >> after there is consensus in the mobility-related WGs
> > about what
> > > > > >> specifically is needed.
> > > > > >>
> > > > > >> Does this sound reasonable?
> > > > > >
> > > > > > It sounds totally reasonable.
> > > > >
> > > > > Good.
> > > > >
> > > > > John
> > > >
> > > > --
> > > > julien.bournelle at int-evry.fr
> > > >
> >
> > --
> > julien.bournelle at int-evry.fr
> >
>