DNS SRV Resource Records for AFS
draft-allbery-afs-srv-records-05

[Ballot comment]
Updated comment after discussion with authors

I support Tim's Discuss and note further that RFC 1183 (which predates
the different classifications of RFCs) says...

  This memo defines five new DNS types for experimental purposes.  This
  RFC describes an Experimental Protocol for the Internet community,
  and requests discussion and suggestions for improvements.

[Ballot comment]
I support Tim's Discuss and note further that RFC 1183 (which predates
the different classifications of RFCs) says...

  This memo defines five new DNS types for experimental purposes.  This
  RFC describes an Experimental Protocol for the Internet community,
  and requests discussion and suggestions for improvements.

Isn't the correct thing to do here to obsolete RFC 1183 rather than
updating some bits and deprecating other bits?

[Ballot discuss]
I will clear on the call after the discussion, but I would like to cosndier whether this document
should be standards track or informational.  It feels informational to me.

[Ballot discuss]
The following sentence (from the end of page 5) needs clarification
"This server ordering MAY be done either per call or for the lifetime of the DNS SRV RR information".

It's not clear what "call" means in this context. I think the document is trying to say that it's an implementation decision on whether to reuse the results of a running the weighted-selection algorithm in 2782. If that's the case, I suggest more guidance to the implementors on when such reuse is appropriate is needed. The load-leveling capabilities the SRV weight field are defeated by reusing the results of the calculation.

If the intent was, instead, to note that the implementation could rerun the weighted-selection algorithm without re-querying DNS, that should be stated more clearly.

[Ballot comment]
What is the [AFS1] Infomative Reference? If it's a book then please mention the publisher. If it's a paper then please mention the source, etc.

[Ballot comment]
Section 4., paragraph 8:
>        _._.
...
>    MUST be the AFS cell name for which the identified server
>    provides AFS services.

  Are AFS cell names required to be valid domain names?

[Note]: 'Jeffrey Hutzelman (jhutz at cmu dot edu) is the document shepherd.
After discussing this with Jari I am thinking that the normative downreference to RFC 1183 (Experimental) is Ok in this case. Note that this downref was explicitly called out in the IETF LC announcement.
' added by Alexey Melnikov

Shepherding write-up:

(1.a)  Who is the Document Shepherd for this document?  Has the
      Document Shepherd personally reviewed this version of the document
      and, in particular, does he or she believe this version is ready
      for forwarding to the IESG for publication?

      >> The Document Shepherd for this document is Jeffrey Hutzelman,
      >> .  I have reviewed this document, and I believe
      >> it is ready for IETF-wide review and publication as a Proposed
      >> Standard.

(1.b)  Has the document had adequate review both from key members of
      the interested community and others?  Does the Document Shepherd
      have any concerns about the depth or breadth of the reviews that
      have been performed?

      >> This document has been reviewed and discussed extensively
      >> among the AFS development community, as well as receiving
      >> review within the IETF.

(1.c)  Does the Document Shepherd have concerns that the document
      needs more review from a particular or broader perspective, e.g.,
      security, operational complexity, someone familiar with AAA,
      internationalization or XML?

      >> No, I have no such concerns.

(1.d)  Does the Document Shepherd have any specific concerns or
      issues with this document that the Responsible Area Director
      and/or the IESG should be aware of?  For example, perhaps he or
      she is uncomfortable with certain parts of the document, or has
      concerns whether there really is a need for it.  In any event, if
      the interested community has discussed those issues and has
      indicated that it still wishes to advance the document, detail
      those concerns here.

      >> No, I have no such concerns.

(1.e)  How solid is the consensus of the interested community behind
      this document?  Does it represent the strong concurrence of a few
      individuals, with others being silent, or does the interested
      community as a whole understand and agree with it?

      >> I believe this document to represent the consensus of the
      >> AFS community, both in its overall intent to introduce use
      >> of SRV records to locate AFS database services and deprecate
      >> use of AFSDB records for this purpose, and in the details of
      >> its specification for how SRV RR's should be applied to AFS.

(1.f)  Has anyone threatened an appeal or otherwise indicated extreme
      discontent?  If so, please summarise the areas of conflict in
      separate email messages to the Responsible Area Director.  (It
      should be in a separate email because this questionnaire is
      entered into the ID Tracker.)

      >> There have been no expressions of discontent.

(1.g)  Has the Document Shepherd personally verified that the
      document satisfies all ID nits?  (See
      http://www.ietf.org/ID-Checklist.html and
      http://tools.ietf.org/tools/idnits/).  Boilerplate checks are not
      enough; this check needs to be thorough.  Has the document met all
      formal review criteria it needs to, such as the MIB Doctor, media
      type and URI type reviews?

      >> The id-nits tool finds no problems with this document.
      >> Additionally, I have verified that the document satisfies
      >> those requirements not checked by the automatic tool.
      >> Note that this document makes references to "AFS", which
      >> is the name of a distributed filesystem product and protocol.
      >> Historically, "AFS" was abbreviation for "Andrew File System",
      >> but that designation has long since been dropped, and today
      >> "AFS" is a proper name with no expansion.

(1.h)  Has the document split its references into normative and
      informative?  Are there normative references to documents that are
      not ready for advancement or are otherwise in an unclear state?
      If such normative references exist, what is the strategy for their
      completion?  Are there normative references that are downward
      references, as described in [RFC3967]?  If so, list these downward
      references to support the Area Director in the Last Call procedure
      for them [RFC3967].

      >> References are properly split.  This document, intended for the
      >> standards track, contains a normative downreference to RFC1183,
      >> an Experimental RFC which defines the AFSDB RR.  The present
      >> document deprecates the use of AFSDB records as describe in
      >> RFC1183 for location of AFS database services, but in the name
      >> of interoperability, recommends that cells advertising AFS
      >> database services via SRV RR's also do so via AFSDB RR's when
      >> possible.

(1.i)  Has the Document Shepherd verified that the document IANA
      consideration section exists and is consistent with the body of
      the document?  If the document specifies protocol extensions, are
      reservations requested in appropriate IANA registries?  Are the
      IANA registries clearly identified?  If the document creates a new
      registry, does it define the proposed initial contents of the
      registry and an allocation procedure for future registrations?
      Does it suggested a reasonable name for the new registry?  See
      [I-D.narten-iana-considerations-rfc2434bis].  If the document
      describes an Expert Review process has Shepherd conferred with the
      Responsible Area Director so that the IESG can appoint the needed
      Expert during the IESG Evaluation?

      >> This document contains a logically empty IANA considerations
      >> section.  Because the names it uses for SRV record service
      >> and proto fields have appeared in the port and protocol
      >> registries for some time, no new IANA actions are needed.

(1.j)  Has the Document Shepherd verified that sections of the
      document that are written in a formal language, such as XML code,
      BNF rules, MIB definitions, etc., validate correctly in an
      automated checker?

      >> This document contains an example zone file illustrating
      >> the use of AFSDB records.  This example zone was successfully
      >> validated using BIND 9.5.0.

(1.k)  The IESG approval announcement includes a Document
      Announcement Write-Up.  Please provide such a Document
      Announcement Writeup?  Recent examples can be found in the
      "Action" announcements for approved documents.  The approval
      announcement contains the following sections:


Technical Summary

  This document specifies how to use DNS (Domain Name Service) SRV RRs
  (Resource Records) to locate services for the AFS distributed file
  system and how the priority and weight values of the SRV RR should be
  interpreted in the server ranking system used by AFS.  It deprecates
  use of the AFSDB RR to locate AFS cell database servers and provides
  guidance for backward compatibility.

Working Group Summary

  This document represents the consensus of the AFS community to
  deprecate the use of DNS AFSDB resource records to locate AFS
  database services, as described in RFC1183, in favor of using SRV
  records.  While the AFS protocols themselves are not the subject
  of any IETF work, this document is being advanced via the IETF
  because it updates previous IETF extensions to the DNS.

Document Quality

  Major AFS client implementors have indicated plans to implement
  support for use of SRV records as described by this document.
  In addition, a variety of developers and operators have indicated
  a desire to publish and use SRV records as described here.  There
  was substantial discussion surrounding the mapping of weight and
  priority information advertised via these records onto the server
  ranking system used by current AFS implementations, which resulted
  in the advice given in section 4.1.

[Note]: 'Jeffrey Hutzelman (jhutz at cmu dot edu) is the document shepherd.
After discussing this with Jari I am thinking that the normative downreference to RFC 1183 (Experimental) is Ok in this case.
' added by Alexey Melnikov

AD review comments were addressed in -03:

Updates: 1183 (if approved)                            December 17, 2009
Intended status: Informational

Should this be Standard Track?


1.  Overview and Rationale

  o  AFSDB RRs do not support priority or ranking, leaving AFS cell
    administrators without a way to control which VLDB servers clients
    prefer.

"clients *should* prefer"?



4.  DNS SRV RRs for AFS

  AFS clients MAY cache targets known to be inaccessible by that client
  and ignore those targets when determining which server to contact
  first.  Clients which do this SHOULD have a mechanism to retry
  targets which were previously inaccessible and reconsider them
  according to their priority and weight if they become accessible
  again.

Maybe this was your intent, but I think this paragraph can make it clear
that any caching needs to obey TTL as specified in the following paragraph:

  DNS SRV RRs, like all DNS RRs, have a time-to-live (TTL), after which
  the SRV record information is no longer valid.  As specified in
  [RFC1034], DNS RRs MUST be discarded after their TTL and the DNS
  query repeated.  This applies to DNS SRV RRs for AFS as to any other
  DNS RR.  Any information derived from the DNS SRV RRs, such as
  preference ranks, MUST also be regenerated based on new DNS query
  results once the TTL for the original DNS information has passed.



5.  Use of AFSDB RRs

  The above is a default recommendation.  AFS cell administrators MAY
  use different lists of servers in the AFSDB RRs and DNS SRV RRs if
  desired for specific effects based on local knowledge of which
  clients use AFSDB RRs and which clients use DNS SRV RRs.  However,
  AFS servers SHOULD NOT be advertised with AFSDB RRs unless they
  provide VLDB and PTS services via UDP on the standard ports.

The last sentence: why SHOULD NOT and not MUST NOT?



8.  References


These need to be split into Normative and Informative.

No Informational reference to AFS?


  [RFC1034]  Mockapetris, P., "Domain names - concepts and facilities",
            STD 13, RFC 1034, November 1987.

  [RFC1183]  Everhart, C., Mamakos, L., Ullmann, R., and P.
            Mockapetris, "New DNS RR Definitions", RFC 1183,
            October 1990.

  [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
            Requirement Levels", BCP 14, RFC 2119, March 1997.

  [RFC2782]  Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS RR for
            specifying the location of services (DNS SRV)", RFC 2782,
            February 2000.

[Note]: 'After discussing this with Jari I am thinking that the normative downreference to RFC 1183 (Experimental) is Ok in this case.
' added by Alexey Melnikov