There is current on-list discussion in the DMARC working group regarding
this issue.

The discussion is centered on whether this is in fact an erratum in RFC
9989, or if it is instead a reason to create a new RFC to replace RFC 9989,
along with a working group chartered to take on that work.

I am personally in the "new RFC and working group to produce it" camp.

RFC 9989 plainly states in section 3.2.13 that for DMARC purposes, a
non-existent domain is consistent with the term's meaning as described in
RFC 8020.

Including RFC 99824's definition and its use of NXNAME seems to expand that
definition in RFC 9989, instead of being a correction of an error therein.


On Wed, Jun 10, 2026 at 2:16 PM <[email protected]> wrote:

> The following errata report has been submitted for RFC9989,
> "Domain-Based Message Authentication, Reporting, and Conformance (DMARC)"
>
> --------------------------------------
> You may review the report below and at:
> https://errata.rfc-editor.org/eid8995/
>
> --------------------------------------
> Type: Technical
> Reported by: Johnathon Mihalop <[email protected]>
>
> Section Appendix A.4. says:
>
> Original Text
> -------------
> If the response code received for a query for a domain name is NXDOMAIN,
> then the name and all the names under it do not exist.
>
> Corrected Text
> --------------
> If the response received for a query for a domain name proves that the
> queried name does not exist, then the name and all names under it are
> considered non-existent for the purposes of this specification. This proof
> can be conveyed by an NXDOMAIN response, or by a DNSSEC-validated
> denial-of-existence response that otherwise indicates non-existence of the
> queried name, such as an RFC 9824 Compact Denial response containing the
> NXNAME type in the relevant NSEC bitmap.
>
> Notes
> -----
> RFC 9824 (Compact Denial of Existence in DNSSEC) is a Proposed Standard
> that defines a way to reduce the size and signing cost of DNSSEC
> denial-of-existence responses. For a non-existent name, Compact Denial can
> return a NODATA-shaped response, i.e. RCODE=NOERROR with an empty Answer
> section, while signalling non-existence using the NXNAME RR type (TYPE128)
> in the NSEC/NSEC3 type bitmap.
>
> RFC 9989 describes non-existent domains only in terms of NXDOMAIN. This is
> incomplete for DNSSEC-validated responses from zones using RFC 9824 Compact
> Denial. A receiver that only treats RCODE=NXDOMAIN as proof of
> non-existence may fail to recognize an RFC 9824 NXNAME denial proof and may
> therefore fail to apply the DMARC np= policy for a non-existent Author
> Domain.
>
> Instructions:
> -------------
> This erratum is currently posted as "Reported". Please
> use "Reply All" to discuss whether it should be verified or
> rejected. When a decision is reached, the verifying party
> will log in to change the status and edit the report, if necessary.
>
> --------------------------------------
> RFC9989 (draft-ietf-dmarc-dmarcbis)
> --------------------------------------
> Title               : Domain-Based Message Authentication, Reporting, and
> Conformance (DMARC)
> Publication Date    : May 2026
> Author(s)           : T. Herr, Ed., J. Levine, Ed.
> Category            : Proposed Standard
> Source              : dmarc (art)
> Stream              : IETF
> Verifying Party     : IESG
>


-- 
Todd
_______________________________________________
dmarc mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to