In message <[email protected]>, Ted Lemon writes:
> On Mar 4, 2014, at 5:58 PM, Mark Andrews <[email protected]> wrote:
> > Please go read draft-andrews-dnsop-pd-reverse.  This technique will
> > work with any zone to be delegated by the ISP.  For most cases the
> > wire is the authenticator by for things like WiMax you will be
> > putting credentials into the box to authenticticate your DHCP request
> > to the ISP as the medium is not secure anyway.
> 
> Your draft is certainly a valid solution to the problem it sets out to =
> solve, but it doesn't actually change the security model I was talking =
> about.   Given that my point was that you have to trust the wire to make =
> this work, I think are in violent agreement at least on this point, =
> except that I would like us to describe a solution that at least =
> addresses the use case where you can't trust the wire.
 
A HMAC-* option for DHCP (c.f. TSIG) would work if we have to
authenticate a DHCP request without relying on any other layers.

-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: [email protected]

_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet

Reply via email to