Hi James, all,
I don't intend spending much more time on this debate - I'll
stop with
noting a few points below.
> -----Original Message-----
> From: James Kempf [mailto:kempf docomolabs-usa.com]
> Sent: Monday, April 23, 2007 11:15 AM
> To: Narayanan, Vidya; Hannes Tschofenig; Vijay
Devarapalli
> Cc: mipshopietf.org
> Subject: Re: [Mipshop] WG Consensus call on AAA based
> handover keys for FMIPv6
>
> Vidya,
>
> >Since then, all the 5 other drafts that were part
of the original
> >consensus call have progressed (to varying
degrees), while
> draft-vidya
> >has been held without explanation. We have
discussed potential
> >conflicts between draft-ietf-mipshop-handover-key
and
> RFC3972 that need
> >to be resolved before that is an acceptable
solution to move
> FMIPv6 to a PS.
> >We need to discuss the implications of RFC3972 on
that
> document on the
> >list and see what needs to be done, before stating
something
> along the
> >lines of "We have already adopted
> draft-ietf-mipshop-handover-key as a
> >WG document for the mechanism based on SeND. This
is sufficient for
> >advancing FMIPv6 as a proposed standard.", as
the email from
> the chairs
> >presently states.
>
> I think everyone, including Rajeev and myself who are
> co-authors on draft-ietf-mipshop-handover-key, agrees
that we
> need to resolve any issues with RFC3972. I took a look
at the
> minutes from the Prague meeting, and I could not make
head
> nor tail of Lakshimanth's comments. I do not know if
that was
> because the recorder couldn't understand what he was
saying
> or was rather because the comments themselves were
expressed
> in a way that made his point difficult to understand.
We are
> now trying to get a clear statement of what
Lakshimnath's
> concerns are, so that we can address them, and we will
> resolve the issue if there is one.
>
As you know, we've had some offline discussion and have some
input from
Tuomas as well on RFC3972 guidelines - I was noting that we
need to have
that discussion and conclude. Anyway, this was not the main
point of my
email.
> In addition, as Rajeev's email and my email here and in
the
> past and my comments at the microphone during the San
Diego
> meeting have supported, everyone is in agreement that
we need
> to adopt an AAA based mechanism. As Vijay has stated,
there
> is no harm in polling the WG periodically about whether
there
> is still interest in doing a work item. This is, in
fact, a
> good idea since it helps determine whether people have
lost
> interest and therefore to weed out items that are not
of
> interest (for example, the low latency work in MIP4 WG,
which
> has lingered there for 5 years with gradually
dwindling, and
> now zero, interest). I think from the response so far I
> believe that there is strong interest, and I hope
draft-vidya
> will soon be submitted as a WG draft. I think this is
> independent of what is needed for advancing FMIP, as
Vijay
> said, but that should not hold up acceptance of
draft-vidya
> because people clearly want it.
>
I was trying to point out the inconsistency in handling a
single
document among the 5 other documents that were part of the
same
consensus call. No reason has really been given for holding
up
draft-vidya subsequent to the consensus call - I simply
don't understand
why it was not adopted just like the other drafts, after
completion of
reviews and necessary updates, when one of the drafts was
adopted even
without the necessary MobDir review!
> Finally, I believe Vijay's email brought out the issue
of
> whether there are any changes needed in the AAA
> infrastructure in order to support draft-vidya. Recall
that
> such changes (though not large) were identified during
the
> bootstrapping design work in MIPv6. So I think it
might be
> useful to consider what would be needed to make
draft-vidya
> work well, and maybe come up with a list of
requirements for
> the appropriate AAA WG if anything is found. Saying
that no
> changes are needed in a misguided competition to
somehow make
> the AAA based mechanism look as or more attractive than
the
> SEND-based mechanism is really not helpful . Not that
anybody
> has made this kind of statement (yet). 
>
Nobody has claimed that no changes are needed to the AAA
infrastructure.
It is known that we need new attributes and we've even had
appendices in
our draft to show that - the appendices were only recently
removed,
since those are supposed to be separate documents anyway
(and all
reviews asked for removal of those). But, a AAA-based scheme
requiring
changes to AAA infrastructure to support additional
attributes/AVPs is
not new.
Vidya
> I think the goal of standardizing draft-vidya for an
AAA
> based mechanism should be to ensure that there is a
> well-designed AAA based mechanism that covers all the
> security issues, with recommended changes to the AAA
> infrastructure if necessary.
>
> jak
>
>
>
>
_______________________________________________
Mipshop mailing list
Mipshopietf.org
https:
//www1.ietf.org/mailman/listinfo/mipshop
|