On 2026/07/21 12:10, Chris Box wrote:
I tried to read the document from the perspective of an unfamiliar
implementer.
Section 3.3 requires an error in the following case.
* An MQTYPE-Query option in a query where the primary question is a
non-data RRTYPE (e.g., ANY, AXFR).
I tried to look up the exact definition of non-data RRTYPE. Domain Name
System (DNS) Parameters <https://www.iana.org/assignments/dns-
parameters/dns-parameters.xhtml#dns-parameters-4> does not contain a
data/non-data column, but implies that 128-255 (among others) are non-
data. It's not clear whether 61440+ are non-data. Looking elsewhere, RFC
6895 <https://datatracker.ietf.org/doc/html/
rfc6895.html#section-3.1> section 3.1 suggests that 41,128-255 might be
the non-data types. I suggest we refer to an appropriate definition so
it's crystal clear what should result in an error.
[with my DNS RRtype Expert Reviewer hat on]
I don't believe there is any clear definition that could be used as a
reference, but to an existing DNS implementor this should be clear. It
is a minor known gap in the DNS specifications that is as yet
unaddressed in the DNS Terminology RFC (9499 and predecessors)
(128 - 255 are indeed non-data, and 41 should've been allocated out of
that range but wasn't. 61440+ are available as data types)
Section 1 is the introduction. My personal feeling is that introductions
should allow the reader to obtain a good idea of the overall concept
fairly quickly. I suggest that the following sentence is a detail that
does not need to be in the introduction, and could instead appear in 3.1.
I disagree. The Intro contains explanatory text, whereas section 3 is
all normative specification language.
Ray
--
auth48archive mailing list -- [email protected]
To unsubscribe send an email to [email protected]
