In your previous mail you wrote:

=> first I have submitted the draft, second I tried to make clear
that my proposal relies only on (any kind of) network access control
(i.e. anything more that just plug and play). AAA is an implementation
option which has extra benefits.

   > - As a general rule, I'd like the Internet to use end-to-end
   >    mechanisms more than network assistance. This isn't just
   >    an architectural principle, but it will also ensure that
   >    we can deploy our things without waiting for providers to
   >    catch up.
   
   I agree.
   But I would personally make a stronger statement since I'm concerned with
   the direction of piling more and more requirements and dependencies on
   AAA. Thus I think the AAA approach is the wrong one - if we collectively
   can make AAA/Diameter do the things needed to make Radius more usable

=> for my purpose Radius is enough (only a new attribute is needed
for the home address, a new version of RFC 3162 ?)

   (reliabiliy, security, some extensibility) I think we've collectively
   have been successful. Keep piling more stuff on the AAA system and it might
   very well get too heavy to be able to fly...
   
=> I agree but AAA is the only way to provide (one day) remote network
access control (*not* necessary but fine).

   Also, waiting for AAA solutions to be available (specified, implemeted,
   and deployed) before MIPv6 can be used seems to be counter to our desire
   to finish up MIPv6 soon.
   
=> I never proposed to wait for AAA solutions (as I ask only for network
access control, not everywhere but enough to make HAO spoofing unattractive).

   While I have concerns with using the AAA per above I think the phased approach
   (where the AAA approach continues to be discussed and further understood) 
   makes sense to me.
   
=> have you still concerns with using network access control in a BCP?

Regards

[EMAIL PROTECTED]

PS (for the IPv6 WG list): my draft is about HAO vs ingress filtering,
I assume there is a consensus about traditional ingress filtering, i.e.
RFC 2827 is applicable to IPv6 (i.e. we don't need another document and
we'll get tradition IPv6 ingress filtering (access lists and unicast RPF)
support in the next router software version (I apologize if this is
already available)).
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to