Re: [v6ops] Please review the No IPv4 draft

[Nick Hilliard <[e-mail]>]

On 16/04/2014 17:42, Ted Lemon wrote:
> On Apr 16, 2014, at 11:10 AM, Nick Hilliard <[e-mail]> wrote:
>> well no, not really.  I can find no discussion at all about this on the
>> sunset4 mailing lists
> 
> There are actually two other drafts related to this topic:

ok, thanks for the refs.  I was aware of draft-yang-sunset4-weaken-dhcp
(and had already looked that up), but not of draft-yang-dhc-ipv4-dis.

> http://tools.ietf.org/html/draft-yang-sunset4-weaken-dhcp-00
> http://tools.ietf.org/html/draft-yang-dhc-ipv4-dis-01

there is no record of any discussion about draft-yang-sunset4-weaken-dhcp
on the sunset4 mailing list; just some notifications that the draft had
been published.

There were several emails on dhcwg@ about draft-yang-dhc-ipv4-dis, only one
of which provided any comments, and another (from you) which contained a
single question which was never answered.

> The authors of draft-yang-sunset4-weaken-dhcp-00 decided to combine
> their work with the current working group draft.   This was discussed in
> the meeting at IETF 87, and there were no objections (although there was
> substantial comment on the draft).   The draft was adopted by the
> working group in September 2013; no objections were raised.

The sunset4 minutes from IETF87 are here:

https://tools.ietf.org/wg/sunset4/minutes?item=minutes-87-sunset4.html

There's no mention of the DHCPv4 issue here either.

There was discussion about this draft in sunset4 at ietf84 through ietf86,
but all the discussion was entitled "Disabling IPv4 with DHCPv6", which
rather put the cart before the horse, so to speak.  Nothing in the minutes
indicated any discussion as to why DHCPv4 was discarded as an option.  This
doesn't mean that there wasn't a discussion, of course.  It just means that
if there was, it wasn't minuted and now we have no way of reviewing the
rationale for the decision.

> So unless you are saying that there was a process failure here, the
> issue is effectively settled.

Process exists to serve the ietf, not the other way around.  Several people
in both v6ops and homenet have expressed serious concern about the choice
to use dhcpv6 and ra instead of dhcpv4.  If you want to invoke process to
close the discussion on this topic, then that's your prerogative, but given
the previous lack of discussion about it, I would view this as a very poor
idea.

Nick