ietf
[Top] [All Lists]

Re: [tsvwg] Last Call: <draft-ietf-tsvwg-byte-pkt-congest-09.txt> (Byte and Packet Congestion Notification) to Best Current Practice

2013-03-07 11:22:03
Thanks for the note. I'd like to respond on the question of intended
status. I hope this may allow others to also comment.

The discussion on the proposed milestone was at  IETF-81, Quebec City,
http://tools.ietf.org/wg/tsvwg/minutes?item=minutes81.html.  This was
followed-up with the ADs, with a result that the document milestone
proposed BCP status.

I do note that this updates an Informational draft, and do not know of
other normative dependencies.

Now is the correct time to check the standards status if this needs to be
reviewed.  As I understand, this is an AD decision.

Gorry
(tsvwg Co-Chair)

The IESG <iesg-secretary(_at_)ietf(_dot_)org> wrote:

The IESG has received a request from the Transport Area Working Group WG
(tsvwg) to consider the following document:
- 'Byte and Packet Congestion Notification'
  <draft-ietf-tsvwg-byte-pkt-congest-09.txt> as Best Current Practice

The IESG plans to make a decision in the next few weeks, and solicits
final comments on this action. Please send substantive comments to the
ietf(_at_)ietf(_dot_)org mailing lists by 2013-03-07. Exceptionally, 
comments may
be
sent to iesg(_at_)ietf(_dot_)org instead. In either case, please retain the
beginning of the Subject line to allow automated sorting.

   This document has a rather long history in TSVWG, and IMHO deserves
to be published as is, but not as a Best Current Practice.

   Fundamentally, it updates RFC 2309 (an Informational document) to
deprecate a practice which seems essentially unused, and goes into detail
appropriate to an academic paper about the theoretical basis for doing
this.

   According to the document writeup I find in the datatracker:
]
] This document is intended as BCP. (This was discussed at IETF-81 and
that
] the status changed from Informational to BCP, because the draft provides
] guidance to implementors and people configuring routers and hosts).

   I cannot find any minutes for that WG meeting. I am willing to believe
Gorry that such a recommendation happened at that meeting (where I was not
present), but I do not find it to have been discussed on-list at all.

   I do not agree that its major purpose is to provide such advice, nor
do I see how "implementors and people configuring" would be likely to
get clear advice from a 43-page document that reads like an academic
paper. For one example:

Abstract
   This document provides recommendations of best current practice for
   dropping or marking packets using active queue management (AQM) such
   as random early detection (RED) or pre-congestion notification (PCN).
   We give three strong recommendations: (1) packet size should be taken
   into account when transports read and respond to congestion
   indications...

   "Packet size should be taken into account when transports read and
respond to congestion indications" is simply too vague. There has been
on-list discussion of what this might mean; but it has not resulted in
clear, concise advice to implementors.

   In no sense do I believe it worth holding up this document any longer
to add "clear advice" -- I believe that would only add years to the delay.
The document deserves to be published, but with Informational status so
folks don't spend their time trying to interpret its "advice".

--
John Leslie <john(_at_)jlc(_dot_)net>



<Prev in Thread] Current Thread [Next in Thread>