Re: [DNSOP] Benjamin Kaduk's Yes on draft-ietf-dnsop-dns-zone-digest-13: (with COMMENT)

["Wessels, Duane" <[e-mail]>]

> On Oct 11, 2020, at 9:03 PM, Benjamin Kaduk via Datatracker <[e-mail]> wrote:
> 
> Benjamin Kaduk has entered the following ballot position for
> draft-ietf-dnsop-dns-zone-digest-13: Yes
> 
> 
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> Thanks for addressing my discuss (and comment!) points.  There are still
> a few more threads to tidy up, but I'm happy with the direction we're
> going.
> 
> Section 1
> 
> We (implicitly) mention "integrity" here as provided in the absence of
> DNSSEC, but later in Section 1.1 we say that integrity can only be assured
> when the zone is signed.  I leave it to Roman to say when his discuss is
> resolved, but it seems likely that we should be consistent about which way
> we go with it.

Looks like I missed that spot in when addressing Roman's point.  Now changed
to this:

   It allows a receiver of the
   zone to verify the zone's integrity and authenticity when used in
   combination with DNSSEC.



> Section 1.1
> 
> It's perhaps unusual to follow "the motivation for this protocol" with "a
> secondary motivation"; instead writing "the primary motivation" would reduce
> the surprise at seeing a secondary motivation added later.

Agreed.  This has been changed.


> 
> Section 2.2.2
> 
> This change seems to be a regression?  The value 1 in question is the
> scheme value, not a Hash Algorithm value.  (I would make this a
> Discuss point but I am sure we will get it resolved quickly.)

Oops, I changed that in the wrong place.  Now it says "with Scheme value 1" there
and "with Hash Algorithm value 1" in the next section.


> 
> Section 3
> 
> (nit) Right now the literal reading of "identical" is that the ZONEMD and
> the signature and the denial-of-existence records are identical, which
> is of course nonsensical.  Perhaps adding "to the ones produced by this
> procedure" or similar would reduce the stress for people who habitually
> make sentence diagrams.

Changed to this:

   Implementations that deviate from the
   described algorithm are advised to ensure that it produces ZONEMD
   RRs, signatures, and dential-of-existence records that are identical
   to the ones generated by this procedure.

> 
> Section 4
> 
> I can't tell if there's a duplicate line in the XML source or not, here
> (as an editing leftover), but that's my guess as to what happened.  In
> particular, I'm not sure how one would query for a DS RR *in the anchor*.
> If I'm reading the previous thread correctly we were only proposing to talk
> about querying for (and validating) DS RRs in the parent zone, not the
> anchor (whatever that means).

Yes indeed there was a line duplicated during editing.  Now:

       This is done by examining locally
       configured trust anchors, and, if necessary, querying for (and
       validating) DS RRs in the parent zone.

> 
> Who is going to come to a conclusion on the "[ Maybe remove all the "SHOULD
> report" above and just say this:]"?  (I'd be fine with it, for what little
> it's worth, but I don't think my opinion is anywhere close to the most
> relevant one.)

Both you and Rob asked about this -- the possibility of overly verbose reporting.
I'd like to hear Rob's opinion.

DW