Paul Vixie wrote:
> Petr Špaček wrote on 2022-07-26 06:21:
> > On 28. 06. 22 16:20, Bob Harold wrote:
> >  > But the parent NS set is not covered by DNSSEC, and thus could be
> > spoofed??
> >  > (Wish we could fix that!)
> > 
> > I share your wish.
> > 
> > Does anyone else want to contribute?
> 
> only to fight such a change.
> 
> > Can people here share their memories of why it is not signed? I wasn't
> > doing DNS when this was designed and I think it would be good to
> > understand the motivation before we start proposing crazy things.
> 
> it exists in two places. only one can be authoritative. therefore only one
> can be signed. as olafur said down-thread, RFC103x ought to have used
> different rr types for delegation vs. apex nameserver list. now we live with
> it.

I don't agree that "only one can be authoritative. therefore only one
can be signed." I don't see a prohibition against signing
non-authoritative RRsets in general. RFC 4035 § 2.2 provides:

   For each authoritative RRset in a signed zone, there MUST be at least
   one RRSIG record…

   [Note: The above is not the same as saying that every RRSIG in a
   signed zone must only cover authoritative RRsets.]

   The NS RRset that appears at the zone apex name MUST be signed, but
   the NS RRsets that appear at delegation points (that is, the NS
   RRsets in the parent zone that delegate the name to the child zone's
   name servers) MUST NOT be signed.  Glue address RRsets associated
   with delegations MUST NOT be signed.

In other words, all authoritative RRsets in a signed zone must be
signed, while non-authoritative NS records and glue address records in a
signed zone must not be signed. This section does not say anything
specific about the signing of non-authoritative RRsets in a signed zone
that are not delegation NS records or glue address records. The reason
delegation NS records and glue address records can't be signed is
because it would violate a specific MUST NOT about those kinds of
records, not because those records are non-authoritative.

So I conclude that "authentic" and "authoritative" are not synonymous in
the DNS(SEC) data model. "Authentic" and "authoritative" being distinct
categories of RRsets is consistent with RFC 2181 § 5.4.1, which
provides:

   When DNS security [RFC2065] is in use, and an authenticated reply has
   been received and verified, the data thus authenticated shall be
   considered more trustworthy than unauthenticated data of the same
   type.

Thus it would be possible to distinguish the trustworthiness level of
"[Authenticated] data from the authority section of a non-authoritative
answer" from "[Unauthenticated] data from the authority section of a
non-authoritative answer", and those two trustworthiness levels could
themselves be distinguished from "authoritative data included in the
answer section of an authoritative reply" and "data from the authority
section of an authoritative answer". In practice I don't believe
resolvers bother with the authenticated/unauthenticated distinction for
individual trustworthiness levels and consider DNSSEC "secure" or
"validated" to be a singular, penultimate trustworthiness level separate
from the levels enumerated in 2181.

The RRSIG "Signer's Name" field unambiguously identifies the zone that
the covered RRset belongs to, and is covered by the signature
calculation. So even if a parent zone and a child zone were signed by
the same key, the RRSIG(NS) at the apex of the child zone is
cryptographically distinguishable from the RRSIG(NS) with the same owner
name at the delegation point in the parent zone.

In practice the specific prohibition on signing delegation NS records
and glue address records in 4035 § 2.2 probably enables a bunch of
implementation simplifications (e.g. no need to key the resolver's RRset
cache by zone name as well as owner name, no need to double up on the
trustworthiness levels, etc.). So the historical reasons for why
delegation NS and glue address records aren't signed is probably less
relevant than the present day interoperability issues that such a change
might cause.

-- 
Robert Edmonds

_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to