Hi Ben, On 17 Nov 2026, at 20:03, Ben Schwartz <[email protected]> wrote:
> On Tue, Nov 17, 2026 at 12:08 AM Joe Abley > <[email protected]> wrote: >> Fetching over ADoX does not provide integrity protection; even if you >> authenticate the server you connect to (which might well be complicated) you >> need DNSSEC to validate the integrity of the response. > > I disagree. If I operate a local full resolver, and perform all > recursive queries over authenticated, integrity-protected connections, > then I can be sure of the data's integrity even if the zone is not > DNSSEC-signed. (This is the position advanced in > draft-rescorla-dprive-adox-latest, although that proposal is > outdated.) I think bear in mind that (a) authentication of servers is not always trivial (nameservers often have many names; addresses can be subverted since the routing system is not strictly trustworthy); (b) unsigned data might have been modified on a secondary server, or in transit between primary and secondary (c) an insecure delegation from a parent cannot be trusted (the NS RRSet above the zone cut is unsigned) so the degree of integrity protection provided by DNSSEC and that provided by ADoX are not the same. >>> On the response side, I do agree that we may need an ADoX equivalent >>> of the "AD" bit, but I don't foresee a need for more than one bit of >>> signal (meaning "this response is fully authenticated by some >>> combination of DNSSEC and ADoX"). >> >> I don't think I understand what you mean by this. Are you saying that >> resolvers should no longer validate but should instead trust authority >> servers to assert that validation is not necessary? > > No, not in general. I am saying that there appear to be some clients > that today make use of the "AD" bit (because they presume a secure > connection to a trusted resolver) [1][2]. Most clients use the AD bit. The number of stub resolvers that perform validation themselves is very small. > If we intend to support > this arrangement, then I believe we should extend it with some signal > to indicate ADoX-based authentication. (We could even just overload > the AD bit, but I suspect that would not have consensus.) I'm not sure what ADoX-based authentication means, above. What would an authority server be signalling in a response to a recursive server by setting AD=1? >> Earlier you suggested that four was too small a number to enumerate the >> privacy options in Leon's draft, but here you seem to say that one bit is >> sufficient to cover all the combinations of DNSSEC validation, secure >> transport and certificate validation requirements? > > I am referring here to the response, not the request. The request is > attempting to express a policy preference, but the response has a much > simpler job: to indicate whether the data could have been poisoned. Poisoned or not is a matter of integrity protection. This draft is about privacy protection of the contents of queries and responses. I don't think these are the same. Joe _______________________________________________ DNSOP mailing list -- [email protected] To unsubscribe send an email to [email protected]
