Amended Appeal Re: grow: Last Call: 'Operation of Anycast Services' to BCP (draft-ietf-grow-anycast) (Dean Anderson; 2006-06-14) - 2006-06-14
Appeal - 2006-06-14
Subject: Amended Appeal Re: grow: Last Call: 'Operation of Anycast Services' to BCP (draft-ietf-grow-anycast) (fwd)
Date: Tue, 13 Jun 2006 21:51:18 -0400 (EDT)
From: Dean Anderson <dean@av8.com>
To: iesg@ietf.org
Clarification has been requested by members of the IESG. It seems that it is
not clear what decisions are disputed. To address this confusion, the original
Appeal is amended to include requests for specific action and proceeding by the
IESG.
Requested Action: Four actions are specifically requested:
1) The IESG should annul the decision by David Kessens (Announcement available
at http://darkwing.uoregon.edu/~llynch/grow/msg00532.html) to move the
draft-ietf-grow-anycast document to "Last Call", and return the document
draft-ietf-grow-anycast to the GROW WG with an instruction that this document is
part of or based on a scientific fraud. The IETF Datatracker site lists 5 items
for the draft-ietf-grow-anycast document on June 1, 2006. Most, if not all of
these should be reversed or undone. The item on June 5, 2006 should also be
removed or reversed as necessary.
Some additional evidence has come to light after a discussion with IESG members.
This dicussion can be found on the IESG list, and will be made available at
http://www.av8.net/IETF-watch/IESG/Appeal-draft-ietf-grow-anycast.html
In particular, there are a number of discrepancies and false or out-of-context
statements in the datatracker document written by Kessens: (document available
at
https://datatracker.ietf.org/public/pidtracker.cgi?command=view_comment&id=51723)
Kessens describes Anderson as the only dissenter. But there are other dissenters
who agreed with Anderson, and then dropped out of further discussion:
http://darkwing.uoregon.edu/~llynch/grow/msg00402.html
Kessens describes a solid consensus. But of 7 persons (excluding chairs and
authors) participating in the discussion, 2 dissent. One plainly dissents
strongly, and indicates 6 unresolved issues. So 28% do not support the document
in draft 2. No comment at all is made on draft 3, not even from the chairs. In
the meantime, the research supporting stateful anycast is found not to perform
stateful testing. Kessens is aware of this fact from the 3/8/06 Appeal.
Kessens statement in the datatracker document seems to be an exaggeration if not
an affirmative misrepresentation. In light of the charges and evidence presented
against Kessens in the Appeal registered 3/8/06 with the IESG for participation
in the fraud, this comment cannot be lightly dismissed as benign. In any case,
5 people do not seem to make a "solid consensus" nor constitute adequate
consideration for a scheme with such global and far reaching effects as Stateful
Anycast, as sought by ISC and RIPE for use on DNS root servers.
Kessens asserts that Anderson does not strongly object to this document. In view
of the assertions made in the Appeal registered 3/8/06 with the IESG, this
statement can only be viewed as an outright lie, or "affirmative
misrepresentation". Kessens writes:
There was a single dissenting view from Dean Anderson noting a
difference in semantic interpretation of 'node selection',
differences in capabilities of 'Per-Packet Load Balancing', and
differences of view in the potential for TCP failure over parallel
diverse paths with Per-Packet Load Balancing. Dean noted that the
chairs had judged WG consensus for publication of this document and
that he did not dispute that judgment.
This is not an honest characterization of Anderson's issues as Anderson
describes them in
http://darkwing.uoregon.edu/~llynch/grow/msg00462.html
http://darkwing.uoregon.edu/~llynch/grow/msg00426.html
For example, the issue with the term 'node selection' is not one of semantic
interpretation, but of a complete fiction of the notion. There is no such
notion: Anycast clients cannot select the anycast instance or node.
The claim that Anderson noted "the chairs had judged WG consensus" was taken out
of context. Anderson was responding to what Huston had
previously written: http://www.uoregon.edu/~llynch/grow/msg00461.html
December 1, 2005 Geoff Huston writes:
==================================================
I have seen no further working group postings on this draft during this
last call period, and no observed working group consensus to create another
revision of this draft.
==================================================
So, I (Anderson) was under the impression that Huston had made a decision and
was not interested in any further discussion, even though I (Anderson) had
documented 6 unaddressed problems.
But it is clear now that in fact no decision had been made in December, and I
(Anderson) was wrong to think it had been. That is not an "admission of
decision".
On January 27, 2006, another draft was issued, according to the actual group
consensus which was to create another draft.
Then we found that the assurances given by ISC were false and based on false
research. Kessens was aware of this discovery.
(See IESG email discussion, made available at
http://www.av8.net/IETF-watch/IESG/Appeal-draft-ietf-grow-anycast.html)
2) The IESG should annul the decision by GROW WG Chair Geoff Huston to assert
that the document draft-ietf-grow-anycast was ready for Last Call, as there were
6 un-addressed issues listed by Anderson on December 2, 2005. There is no
comment on the new draft that was issued January 27, 2006 to indicate consensus.
Also the Chair made an incorrect technical choice, since the subject matter of
the document doesn't work, as outlined by Anderson, previously.
3) The IESG should take appropriate action against David Kessens and Brian
Carpenter for their roles and multiple acts in perpetrating a scientific fraud
of stateful Anycast.
4) The IESG should take appropriate action with respect to other inappropriate
or unethical activities that are brought to light as a result of investigation,
including unethical or inappropriate attempts to influence or interfere in this
or other proceedings.
Proceeding:
Brian Carpenter should proceed by recusing himself from this matter, and halting
attempts to interfere or influence this proceeding. The IESG should elect a
temporary Chair to oversee its proceedings according to is usual process if
there is a usual process when the Chair is absent.
David Kessens should proceed by recusing himself from this matter, and halting
attempts to interfere or influence this proceeding.
The IESG should proceed as though Brian Carpenter and David Kessens are absent
and should conduct its considerations without interference or influence from
Carpenter or Kessens. Evidence of interference or influence should be presented
as cause for action.
Original Appeal:
---------- Forwarded message ----------
Date: Fri, 2 Jun 2006 16:07:47 -0400 (EDT)
From: Dean Anderson <dean@av8.com>
To: iesg@ietf.org
Cc: grow@lists.uoregon.edu
Subject: Appeal Re: grow: Last Call: 'Operation of Anycast Services' to BCP
(draft-ietf-grow-anycast)
I object to this document on the following grounds:
Objections were made to the DNSOP WG regarding at least 6 problems with the
document that were not addressed:
http://darkwing.uoregon.edu/~llynch/grow/msg00426.html
http://darkwing.uoregon.edu/~llynch/grow/msg00462.html
There may be additional problems.
The notion of a safe stateful anycast operation as asserted by Daniel Karrenberg
(http://www.nanog.org/mtg-0505/pdf/karrenberg.pdf) has now been discredited.
Karrenberg's document misled people to believe that stateful anycast was safe,
when in fact Karrenberg didn't perform any stateful testing whatsoever.
So, there is no evidence that stateful anycast is safe, and substantial evidence
that it is not safe: Mark Kosters reports on data gathered at J root:
http://www.nanog.org/mtg-0410/pdf/kosters.pdf
+ Expected to see a saw tooth distribution .
instead have a noisy distribution in many cases
+ Does not affect UDP
+ DO NOT RUN Anycast with Stateful Transport
http://www.rssac.org/meetings/04-08/2004WashDC.html
Kosters repeats warning on stateful DNS Anycast, but is disputed by
Karrenberg. It is later found (January, 2006) that Karrenberg has done no
stateful testing whatsoever, and did not reveal that his testing was only for
stateless DNS, and therefore irrelevant to Kosters data. This discovery was
only made when Anderson examined the source code to the DNSMON program written
by Karrenberg to conduct this testing.
Stateful transport includes large UDP ENDSO packets which are required by
DNSSEC.
The draft-ietf-grow-anycast document incorrectly gives the impression that
stateful anycast is safe. This is a an incorrect conclusion based on
discredited research. Therefore this document should not be accepted for
technical reasons.
DNS Anycast controversy and inappropriate behavior by officials during
discussion of this document gives rise to questions on both the integrity of the
document, and questions as to whether all points of views have been adequately
considered.
To summarize the controvery: During discussion of this document and the subject
of DNS Anycast and DNS Root Anycast, David Kessens, Area Director for the
Operations Area which includes DNSOP WG and the GROW WG, attempted to conspire
with Brian Carpenter, David Crocker, and Susan Harris to improperly silence
discussion of problems with DNS Anycast [and the same persons also tried to
improperly silence questions regarding the integrity of an IETF spam document
authored by Crocker]. Then 4 IESG members acted with conflicts of interest in
violation of the ISOC and IETF charter and rules to silence discussion of this
matter. An IAB appeal documents this inappropriate behavior. The IAB has not yet
ruled on the matter.
For example, during this time, Kessens asserted (incorrectly) that DNS Root
server operations were off-topic for the DNSOP WG, and then inappropriately
demanded that discussion of DNS Root Anycast on the DNSOP WG be halted.
Subsequently, it was proposed that DNSOP WG be re-chartered to remove DNS root
server operations from its charter. Somewhat strangely, but consistent with
other absurd allegations, Kessens et al refused to concede that DNS Root Anycast
was presently on-topic for DNSOP WG. There are other allegations too numerous to
fully list here. See http://www.iab.org/appeals/index.html "Appeal Against IESG
PR-Action from Dean Anderson, 18 April 2006" for more information.
As a result, my views have not been adequately presented or considered.
An appeal is therefore registered under Section 6.5.1 and 6.5.2 of RFC 2026.
Dean Anderson
Av8 Internet, Inc
I have begun to collect a history of DNS Anycast at
http://www.av8.net/IETF-watch/DNSRootAnycast/History.html
On Fri, 2 Jun 2006, The IESG wrote:
The IESG has received a request from the Global Routing Operations WG to
consider the following document:
- 'Operation of Anycast Services '
as a BCPThe IESG plans to make a decision in the next few weeks, and solicits
final comments on this action. Please send any comments to the
iesg@ietf.org or ietf@ietf.org mailing lists by 2006-06-16.The file can be obtained via
http://www.ietf.org/internet-drafts/draft-ietf-grow-anycast-03.txt
web user interface: http://darkwing.uoregon.edu/~llynch/grow.html
web archive: http://darkwing.uoregon.edu/~llynch/grow/
--
Av8 Internet Prepared to pay a premium for better service?
www.av8.net faster, more reliable, better service
617 344 9000