On 9 October 2016 at 21:32, Mark Andrews <ma...@isc.org> wrote:
> In message <CAAiTEH9Rbw4u3gV9GULQ-8WdoPHf3SXQMTRY+CtfUGrNQSGAWw@
> mail.gmail.com>, Matthew Pounsett writes:
> > My first impression of this document is that it is still in need of some
> > extreme editing – mostly for grammar and syntax, but also for clarity and
> > readability. I've included many of the early problems I found in a list
> > of nits at the end of this email, but at two and three errors per
> > (and occasionally per sentence) it's just too much to cover this way, so
> > stopped noting nits after the first couple of sections. I suggest such
> > sweeping changes to later sections that any nits I noted there may end up
> > being rewritten anyway.
> > The draft still suffers from inconsistent approaches to the content. In
> > any given section it might shift from a problem description, to
> > admonishment, to advice for workarounds, and back again without a clear
> > consistent structure. This makes it difficult to figure out what the
> > intended takeaway is in places. I've tried to point out specifics below,
> > and suggest better approaches where I can.
> > Don't let my criticism of the document give you the wrong impression. I
> > think this is useful work, and I'm very appreciative of the work that has
> > gone into it so far. I just happen to think that there is still work
> > to be done.
> > As for the meat of the matter...
> > In the Introduction, in the paragraph which begins "Unless a nameserver
> > under attack" would it make sense to replace "broken delegations" with
> > "lame delegations"? We have a term for delegating to a name server
> > is not authoritative for the delegation... we should probably use it.
> > Section 2, "Consequences," seems out of place to me. It makes a number
> > assertions that seem to follow no particular flow, or pattern. The line
> > of
> > reasoning or evidence that lead to these assertions has not appeared yet
> > in
> > the document, and so they seem unsupported. If I'm interpreting the
> > author's intent correctly, perhaps this section should be split up and
> > interspersed into the current section 3 as the direct or indirect
> > consequences of each type of dropped response.
> > The way section 3 is titled and introduced I expect to just see a survey
> > of
> > commonly dropped queries, but the content seems to be inconsistent as to
> > whether it's just a survey, or a survey and advice to client authors
> > working around problems, or just advice without a clear description of
> > problem behaviour. It seems to me that if the section is intended to be
> > advice to authors, then it should be introduced that way. Also,
> > clarification for server authors on how they should respond–instead of
> > dropping queries–should probably come before any advice on how to work
> > around dropped queries.
> > If the goal here is to help implementors fix what's broken, then I would
> > suggest each subsection start with a clear description of an observed
> > broken behaviour, followed up with a description of how that behaviour
> > should change (possibly only as a short summary with external references,
> > as we don't want to reproduce whole RFCs here) and then follow that with
> > advice for client authors for working around existing brokenness.
> > Sticking
> > to a pattern–either the one I've suggested or some other–for every query
> > type would make the entire section easier to read. It will also make it
> > easier for the working group to go through the section item by item and
> > discuss whether we agree on the problem descriptions and proposed
> > remediation.
> > RFC 2119 language is conspicuously absent from this section and SHOULD be
> > added.
> > Section 4, "Remediating", is very problematic. I think the biggest
> > stem from two false assertions:
> > 1) "TLD operators are being asked to do this as they...have access to a
> > large numbers of nameserver names as well as contact details for the
> > registrants of those nameservers."
> > This is incorrect. Or, at least, frequently incorrect. TLD registries
> > fall into one of two categories: thick or thin. Thick registries do
> > this information, but a significant number of thin registries exist,
> > delegate the responsibility for maintaining registrant contact
> > to the registrar. All thin registries know is that a delegation should
> > exist, which name servers are currently supposed to be authoritative for
> > the subzone, and which registrar asked them to create the delegation.
> > This false assertion leads the author down the path of making the TLD
> > registry responsible for contacting registrants. The third paragraph
> > backpedals a bit on the question of who should be doing what, but then
> > rest of the section goes on to talk about the registry anyway.
> > A lot of text could be removed from the first few paragraphs by removing
> > the need to soften the initial statement that this responsibility lies
> > with the TLDs.
> > 2) The second false assertion is the implied assertion that RFC 1033 says
> > anything at all about a zone operator's responsibility to remove
> > delegations that lead to badly behaving name servers.
> > To provide context ... the title of RFC 1033 is "DOMAIN ADMINISTRATORS
> > OPERATIONS GUIDE," and the first sentence of the referenced COMPLAINTS
> > section is, "these are the suggested steps you should take if you are
> > having problems that you believe are caused by someone else's name
> > server."
> > The specifically referenced step is to "ask the parent authorities to
> > excommunicate the domain." None of this is normative language directing
> > parent operator to do anything.
> So we have a peer review document, the author of whom worked for
> the company that managed the COM and ARPA zones and others, saying
> as the step of last resort "request excommunication" and that it
> would not happen if it was justified? I know when I got delegations
> back around '88 that excommuniction for failure to maintain a standards
> compliant server was a real possibility. Unfortunately the email
> that said this is long gone.
Who the author was doesn't change what they wrote.
> > RFC 1033 appears to be nothing more than a "how to" guide for new DNS
> > operators of the late '80s. Its main goal seems to be to give a
> > prospective operator an overview of the DNS, and instructions for how to
> > get a DNS server up and running for the first time. We shouldn't be
> > looking to it for standards guidance any more than we should look to an
> > early (or any) edition of DNS & BIND.
> Actually it has a lot more relevence that DNS & BIND as it was
> written by those that were in control at the time. You will note
> that the author was from SRI and SRI would act on excommunication
> requests if they felt they were justified.
SRIs policy in the 80's doesn't have any bearing what is written in 1033.
Nor do their policies in the 80's have any bearing on the world as it
> > Even if the normative language implied in this draft existed in an RFC
> > somewhere, it is legally impossible for most TLDs to follow the
> > this draft gives. Any registry that falls under the "gTLD" category is
> > contractually prohibited from directly contacting its registrants, and
> > none
> > of the registries or registrars for gTLDs would have a legal leg to stand
> > on if they tried to remove a delegation due to a misbehaving name server.
> > Even if the name server belonged to the registrant of the delegation this
> > would be problematic, but in many cases the registrant and the DNS
> > operator
> > are not the same. Attempting to remove a registrant's delegation because
> > of the actions of a third party would have any gTLD's legal department up
> > in arms in an instant.
> But not all registries as so constrained. This is BEST current
> practice not LOWEST COMMON DEMONINATOR practice.
> GTLD are required to remove records for abuse so removal of record
> is expected for some conditions so it is not beyond belief that
> they can change to include these reasons. ICANN still has a committee
> to maintain the stability of the DNS. Nameserver behavior very
> much affects that stability.
The draft can say it would be helpful to take action. The draft can't
require action. Requiring action isn't describing a best current
practice. That just won't fly on today's Internet.
If you want to lobby ICANN to modify the gTLD agreements, then we can
revisit this discussion. But until those are changed the IETF has no
business insisting on contractually prohibited actions.
> > Removing the above two assertions obviates a lot of the current content
> > section 4, and leaves only descriptions of testing that MAY be
> > implemented,
> > but is already discussed in more detail in section 9.
> > If this section remains, removal of any suggestion of imperative from it
> > is
> > essential to this document being viable.
> > Section 5, "Firewalls," seems to be more of section 3 and should be
> > there. If the author wanted to separate the misbehaviour of name servers
> > from the misbehaviour of middle boxes that would be reasonable, but then
> > this should at least appear immediately after section 3, without some
> > other
> > subject sitting between them.
> > Same for section 6. This just seems to be a special case of the firewall
> > problem.
> > And again for section 7. This definitely belongs in section 3 as just
> > another case of a misbehaving name server.
> > The third paragraph of section 8 fails to address recursive servers
> > handling unknown types. This may just be an oversight if the intent was
> > to
> > limit the paragraph to authoritative servers. If that's the case, the
> > first sentence of the paragraph should be rephrased. It's important to
> > set
> > the reader's expectations for scope as early in the subject as possible.
> > Section 8 as a whole probably belongs closer in the document to section
> > as it is entirely advice for implementors. It could probably be split up
> > into section 3 in much the same way I suggested with section 2.
> > Nits:
> > – Introduction:
> > "Failure to respond to a query is indistinguishable from a packet loss."
> > This seems awkward to me. I'd change it to either "from packet loss" or
> > "from a lost packet."
> > In the following sentence, "a analysis of query response patterns will
> > results"
> > contains two errors. It should be "an analysis of query response
> > will result..."
> > "Servers which fail to respond to queries to remain results in developers
> > being hesitant to deploy new standards." I'm not sure what "to remain"
> > is
> > doing in there. Should this be: "Servers which fail to respond to
> > queries
> > result in developers being hesitant to deploy new standards." ?
> > – Consequences
> > "Lack of following the relevant RFCs has lead to various consequences.
> > Some as a direct result and some from recursive servers try to work
> > the non compliance." This is a sentence fragment; replace that period
> > with
> > a comma.
> > "Fixing known issues know" Fixing known issues now?
> > "Wide spread non response to EDNS queries has lead to recursive servers
> > having to assume EDNS may not supported and fallback to plain DNS is
> > required." I think that "is" is a typo of "as". This also reads like
> > fallback is another thing they must assume. May I suggest: "Widespread
> > non-response to EDNS queries leads recursive servers to assume EDNS may
> > not
> > supported, and fallback to plain DNS as a result."
> > "3. Common queries kinds that result in non responses" ... "Common kinds
> > of
> > queries that result in non response"?
> Mark Andrews, ISC
> 1 Seymour St., Dundas Valley, NSW 2117, Australia
> PHONE: +61 2 9871 4742 INTERNET: ma...@isc.org
DNSOP mailing list