Re: [TLS] Eric Rescorla's Discuss on draft-ietf-tls-dnssec-chain-extension-06: (with DISCUSS and COMMENT)

[Shumon Huque <[e-mail]>]

On Wed, Feb 21, 2018 at 2:48 PM, Paul Wouters <[e-mail]> wrote:

> On Wed, 21 Feb 2018, Shumon Huque wrote:
>
> Okay, got it. For people that have already implemented this, I think
>> there has been an implicit understanding that the format of the chain
>> data is a sequence of concatenated wire format RRs exactly as they appear
>> in a DNS message section
>>
>
> Note, I would not call it "sequence of concatenated wire format RRs", as
> it is simply the wireformat of a DNS message.

The fact that some parts
> are concatenated RR's is just part of the wireformat of the DNS reply
> message. That is, this isn't introducing some new form of concatenated
> content - it is _just_ a regular DNS wire format message, and the
> document should not go deeper into its explanation.
>

It would _not_ be correct to say that this is a "DNS wire format message" -
that
would mean there is a DNS header section with flags and response codes,
and other sections. The chain data structure as currently specified really
is
concatenated wire-format RRs (as they appear _within_ a DNS message
_section_). Let me know if that is unclear in the draft (or my proposed
edits).

I recall at one point way back there was a discussion about whether the
chain
data should just be a fully formed DNS message, but I don't believe that
idea
had much support in the working group (personally I would have been fine
with
that choice too, if it did have support).


> I noticed some discussion about the ordering of this content. I am not
> sure why that should be done. DNS doesn't care about the order, and
> neither should producers or consumers of this extension. DNS has no
> ordering inside its message.
>

Yes, that is in fact where we ended up, roughly.

The draft does currently say that the answer record(s) appear first. And
then
the authentication chain records follow and that TLS client should be
prepared
to receive the authentication chain records in any order.

Requiring the answer records first seems logical to me - it is what
existing
libraries do, and also what the EDNS chain query spec does (answer records
appear first in the ANSWER section, and DNSSEC authentication chain records
appear in the AUTHORITY section in any order).

There is some residual wording in the draft about ordering of CNAMEs etc in
the answer records part. Assuming the WG agrees, I am fine with relaxing
that requirement - that ended up in there because although there is no
defined
ordering of RRs within a DNS message section, as a practical matter CNAMEs
are almost always ordered since there are some DNS queriers that get
confused
otherwise. This is a new protocol though, so we can be more faithful to the
DNS
spec.

I don't think getting unrelated DNSSEC records would be an issue. TLS
> has its maximum sizes for the handshake. In fact, it could allow the
> extension to have some useful data in the case of MX or SRV.  (and could
> be a feature to build a nice resolver over TLS using 1 tor circuit).
>

The draft currently doesn't address the MX or SRV use case. I suggest
that we tackle that in a future version.

I'm also not sure about the talking of unsigned CNAME records from
> DNAME. The above pseudo code (extended with special cases) should be
> in some DNS library, and that library will know what records to expect
> unsigned which are proven by the DNAME (or wildcard) synthesis and knows
> when/if to add it to the validated cache. I don't think that should be
> explained in this RFC at all. The DNS implementation does not need
> to be specified in this document and it should just focus on saying
> that "the DNS message response is validated and upon validation the
> content can be considered DANE validated".


Where we ended up, is that WG participants asked for some level of DNSSEC
detail to be included in this doc.

Shumon.