Early morning fat-fingering - my e-mail should have read "Dave Allan"...
- Ralph
-----
I exchanged e-mail with Davw Alan (DSLF liaison to IETF) at the end
of last week about this discussion of the DSLF liaison statement.
Dave posted a note to the appropriate DSLF mailing list making DSLF
members aware of the int-area lis discussion and asking them to
participate and provide DSLF input.
- Ralph
On Oct 21, 2007, at Oct 21, 2007,9:44 AM, Iljitsch van Beijnum wrote:
On 19-okt-2007, at 22:47, Ralph Droms wrote:
Reading the latest liaison statement, starting with the title "DSL
Subscriber Authentication using DHCP" and the request for the IETF
to take on draft-pruss-dhcp-auth-dsl and/or draft-zhao-dhc-user-
authentication-02 as a work item, leads me to believe the DSLF is
asking now for a DHCP solution rather than, more generally, an L3-
based access authentication protocol. If the int-area discussion
leads to consensus that DHCP is not the right basis for
"subscriber authentication", I suspect we will need to build a
solid case arguing for some other solution in our response to the
DSLF.
Speaking as an individual IETF contributor, I don't think we have
enough information to make a good judgment about whether to adopt
a DHCP-based solution as a work item, or whether some other
standards can be composed into a solution.
I'm not sure what the normal procedure is in these cases, but it
seems advantageous to me to make the DSL Forum aware of the content
of our discussion in the past weeks. Based on that, the DSL Forum
may decide it's a good idea to explore PANA or 802.1x. Or they may
want to stick to a DHCP solution, in which case I'm very interested
in learning their opinion about the difficulties with modifying
existing DHCP implementations and how to handle IPv6.
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
