Re: [DNSOP] Second Working Group Last Call - draft-ietf-dnsop-nsec-aggressiveuse

[Stephane Bortzmeyer <[e-mail]>]

On Tue, Dec 20, 2016 at 07:38:08PM +0000,
 Warren Kumari <[e-mail]> wrote 
 a message of 72 lines which said:

> > * synthesis of NXDOMAIN from NSEC (obviously; that's the minimum)
> > * synthesis of NXDOMAIN from NSEC3 (if no opt-out)
> > * synthesis of NODATA from NSEC/NSEC3
> > * synthesis of positive answers from wilcards+NSEC
> > * all of them?

> The Google Public DNS code is constantly evolving - I'm discussing with the
> team lead to see what answers I can provide to the above....

:-(

> Is this a "nice to know", or do you think it needs to hold up the
> WGLC? Can / should I just remove the section?

To me, it is useful: the goal of this section (RFC 7942) is to inform
people about whether the idea has been tested on the battlefield or
not.

Also, it may help address the remarks by JINMEI, Tatuya
<https://mailarchive.ietf.org/arch/msg/dnsop/lCEW5hef9zmMjKSX4J-KlM3mStY>
(about the fact that NODATA synthesis has not been really seriously
studied).

The fact that it mentioned Unbound for several iterations of the draft
while Unbound actually does not implement the draft seems to indicate
that IETF is not careful enough about "running code" :-(