On 1/15/2024 2:20 PM, Warren Kumari wrote:
These include two Errata filed by Bernie Hoeneisen (author of RFC6118)
against RFC8552 - "Scoped Interpretation of DNS Resource Records
through "Underscored" Naming of Attribute Leaves"
<https://datatracker.ietf.org/doc/rfc8552/>
I'd like to get feedback by Monday Jan 29th, otherwise I'll just go
with my proposed resolutions below.
Errata1: https://www.rfc-editor.org/errata/eid7066
---
Section 4.1.2. says:
| URI | _dccp | [RFC7566] |
It should say:
?
Notes:
Wrong reference. RFC7566 does not even mention "dccp". Unaware of a
useful reference. Not sure this line needs to be removed.
Note that this also has an impact to the IANA registry:
https://www.iana.org/assignments/dns-parameters/dns-parameters.xhtml#underscored-globally-scoped-dns-node-names
—-
I **think** that the correct reference for this is actually RFC7553 -
"The Uniform Resource Identifier (URI) DNS Resource Record"
<https://datatracker.ietf.org/doc/rfc7553/>. Note that this initially
confused me, because I was looking for DCCP (and TCP and UDP) in the
URI *registry*, not in the DNS RR definition.
My proposed resolution: Mark Errata verified, update reference to be
RFC7553.
I am not finding the string dccp in rfc7553.
I looked in the other candidate RFCs and didn't find _dccp in them.
Absent affirmative knowledge that this _attribute is in real-world use,
the correct action would be to remove it from the registry, IMO.
In the alternative -- since it is not exactly damaging or wasting a
precious resource -- leave the registration but take out the reference.
So it shows as registered, but implies it is there because, well, we
felt like it.
On reflection, quite a few of the entries were, I think, done for
exactly that reason. Or rather, for completeness. Note, for example,
that the SRV registration for _dccp points to the SRV RR definition,
although that document does not cite dccp.
Errata 2: https://www.rfc-editor.org/errata/eid7068:
—-
Section 4.1.2. says:
| URI | _tcp | [RFC6118] |
| URI | _udp | [RFC6118] |
It should say:
?
Notes:
Wrong reference. RFC6118 does not even mention "tcp" nor "udp".
Unaware of useful reference(s). Not sure this line needs to be removed.
Note that this also has an impact to the IANA registry:
https://www.iana.org/assignments/dns-parameters/dns-parameters.xhtml#underscored-globally-scoped-dns-node-names
---
My proposed resolution: Same as above.
I think that these, two, were cases of wanting completeness in the
registrations.
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net
mast:@dcrocker@mastodon.social
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
