[AFS3-std] Fwd: future of the AFSDB RR related IANA registry

[Russ Allbery]

--=-=-=

Forwarding this along to reach more of the AFS community.  I wrote back
and said that I agreed with the closing of the registry and would be very
surprised if anyone objected or anticipated further AFSDB record type
registrations, but that I'd pass along any objections.


--=-=-=
Content-Type: message/rfc822
Content-Disposition: inline

From: Alfred =?utf-8?Q?H=C3=B6nes?= <ah@TR-Sys.de>
Message-Id: <201203142135.WAA10471@TR-Sys.de>
Subject: future of the AFSDB RR related IANA registry
To: rra@stanford.edu
Date: Wed, 14 Mar 2012 22:35:58 +0100 (MEZ)
Cc: d3e3e3@gmail.com (Donald E. Eastlake 3rd)
X-Mailer: ELM [$Revision: 1.17.214.3 $]
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Russ,
you might recall that we were in contact during the draft
stage of your RFC 5864, "DNS SRV Resource Records for AFS",
regarding DNS specifics and some editorials.

That RFC already says in its Abstract,

  "This document ... ... ...                        [It] updates RFC
   1183 to deprecate the use of the AFSDB RR to locate AFS cell database
   servers and provides guidance for backward compatibility."

This is further elaborated upon in the body of the RFC.

You also might be aware of the existence of the IANA Registry named
"Domain Name System (DNS) Parameters", which contains a sub-registry
entitled "AFSDB RR Subtype".  Currently, the IANA policy for that
sub-registry is specified in Section 3.1.4 of BCP 42, RFC 6195.

In the DNSEXT community (WG is bound to be closed), there is a
project underway to revise that RFC, and Donald Eastlake (copied)
already has sumbitted an rfc6195bis draft.  I have suggested to also
re-consider the rules for the "AFSDB RR Subtype" sub-registry at
this opportunity, based on my knowledge of your AFS SRV usage RFC.

That sub-registry has, since its conceptional inception by RFC 1183,
never seen any IANA assignment of new code points, and I very
strongly suspect that the AFS community will not see the need or
desire to get any new code points assigned there any more.
Therefore, I have proposed to declare that sub-registry as "closed"
in the future update to RFC 6195.  It could also be tagged as
"Historic", with a pointer to RFC 5864.

This message serves to poll the AFS community whether such action
makes sense and is agreed upon, using you as a kind of (inofficial)
"liaison" to the AFS community.

You may forward this message (or suitable excerpts of it)
under the IETF "Note Well" conditions to third parties
that you want to contact/poll to this end.


For your convenience, here is -- slightly beautified for clarity
and better fit into email and draft text -- the current snapshot
of the sub-registry:


-------- snip --------

Registry Name: AFSDB RR Subtype

Reference: [RFC6195] [RFC1035]

Registration Procedures:

Range
Decimal      Hexadecimal    Registration Procedures
-----------  -------------  --------------------------------------
0            0x0000         Standards Action
1-65279      0x0001-0xFEFF  IETF Review
65280-65534  0xFF00-0xFFFE  Private Use
65535        0xFFFF         Standards Action

Registry:

Decimal      Hexadecimal    Description                       Reference
-----------  -------------  --------------------------------  ---------
0            0x0000         Reserved                          [RFC6195]
1            0x0001         Andrews File Service v3.0         [RFC1183]
                            Location Service
2            0x0002         DCE/NCA root cell directory node  [RFC1183]
3-65279      0x0003-0xFEFF  Unassigned
65280-65534  0xFF00-0xFFFE  Reserved for Private Use          [RFC6195]
65535        0xFFFF         Reserved                          [RFC6195]

-------- snip --------


I envision the future content of it might look like this:


-------- snip --------

Registry Name: AFSDB RR Subtype

Reference: [RFC6195bis] [RFC5864] [RFC1183] [RFC1035]

Registration Procedures:

  This registry is closed after RFC 5864 has deprecated the AFSDB
  Resource Record, and hence no more registrations shall be allowed.
  (Note that since its inception per RFC 1183, no additional code
  points had been assigned.)

  For the historical record, this was the previous assignment policy:

Range
Decimal      Hexadecimal    Registration Procedures
-----------  -------------  -----------------------
0            0x0000         Standards Action
1-65279      0x0001-0xFEFF  IETF Review
65280-65534  0xFF00-0xFFFE  Private Use
65535        0xFFFF         Standards Action

Registry:

Decimal      Hexadecimal    Description                       Reference
-----------  -------------  --------------------------------  ---------
0            0x0000         Reserved                          [RFC6195]
1            0x0001         Andrews File Service v3.0         [RFC1183]
                            Location Service
2            0x0002         DCE/NCA root cell directory node  [RFC1183]
3-65279      0x0003-0xFEFF  Unassigned
65280-65534  0xFF00-0xFFFE  Reserved for Private Use          [RFC6195]
65535        0xFFFF         Reserved                          [RFC6195]

-------- snip --------


I'd appreciate to learn about your opinion on this suggested change.
Alternative suggestions -- preferably with concrete text proposals --
are welcome as well.

Please copy Donald Eastlake (e.g. by "reply to all" or similar)
in your response.


Kind regards,
  Alfred H=C3=B6nes.

--=20

+------------------------+--------------------------------------------+
| TR-Sys Alfred Hoenes   |  Alfred Hoenes   Dipl.-Math., Dipl.-Phys.  |
| Gerlinger Strasse 12   |  Phone: (+49)7156/9635-0, Fax: -18         |
| D-71254  Ditzingen     |  E-Mail:  ah@TR-Sys.de                     |
+------------------------+--------------------------------------------+


--=-=-=--