On 29/07/2008, at 3:52 PM, Bernard Aboba wrote:
> I am sorry but I do not buy that at all. We have implementations
that behave
> this way and we now need to somehow pretend for your academic
reasons that they
> do not exist?
Is the question under discussion whether to document what has
already been implemented?
If so, then discussion of the merits/demerits is indeed academic --
but so is discussion of
adoption of the document as an IETF WG work item. Just put the
documentation on a web
site, or send it to the RFC Editor for publication as an Independent
Submission.
Or is the question whether the document is an appropriate basis for
future work? If so,
then discussion of merits/demerits is not at all academic -- it is a
core issue.
This part of the thread was not about DHCP Authentication but the
relay behaviour in broadband networks. Basically current behaviour is
to do a Radius Access Request on the DHCP discover and fill in
information from the Access Response into the DHCP Offer. Now in some
cases IP address comes from the AAA server or the DHCP Discover is
relayed to a DHCP server and then we have the situation that
information from the Access Request can be "inserted" in at what is a
type of a relay function and their was repeatedly expressed concern
from Alper at that.
It has become clear to many folk that this is not written up in the
IETF and in some conversations we had post this mornings WG meeting
Peter Arberg from Redback has kindly offered to start a write it up
from the DSL community so the behaviour of this function can be
documented, independent of the fate of the Subscriber Authentication
proposals.
- Ric
_______________________________________________
Int-area mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/int-area
