On 10/29/2019 5:01 PM, Eric Rescorla wrote:
> Ben,
>
> Is what you're saying here that .com provides the NS record for
> example.com <http://example.com> and that may not itself be
> example.com <http://example.com>, but instead ns.server.invalid, and
> therefore if you can't trust .com then it doesn't matter if
> ns.server.invalid has a WebPKI cert?


We want to assure integrity and confidentiality of the exchanges, which
clearly requires authenticating the server. So if for .com the server
"a.gtld-servers.net" provides for "privateoctopus.com" the NS record
"ns-125.awsdns-15.com", I want to authenticate "ns-125.awsdns-15.com".
That's exactly how it is supposed to work.

The related question is whether "ns-125.awsdns-15.com" is in fact
authorized to provide answers for "privateoctopus.com". ADOT provides
some assurance of that, but if you want proofs the answer is DNSSEC.

But then, there is a special concern for the scenario in which the
authoritative server has been corrupted and sends back a bogus A record
for "www.example.com", pointing to an attacker-controlled copy of the
original server, and then using DNS based verification process to obtain
a bogus PKI certificate for the copy.

In practice, issuance of domain based PKI certificates relies on the
integrity of the DNS. ADOT plays a role in that integrity. It would be
better if the integrity of ADOT did not depend on PKI, because that
would introduce a circular dependency. Using DANE instead of PKI there
seems prudent.

-- Christian Huitema

_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to