Re: [Add] [EXTERNAL] Re: New Version Notification for draft-pauly-add-resolver-discovery-01.txt

[Tommy Jensen <[e-mail]>]

Hey Tim,

One reason we see our draft as a better approach is that we reuse the record type used for advertising the keys needed for Encrypted ClientHello. This means a client wanting to implement both standards only needs to add one new query type rather than two.

Notice our primary discovery mechanism is a DNS approach (querying for the designation record), as well as our bootstrapping mechanism (querying a resolver for its DoH configuration and confirmed during its handshake). I think we agree on DNS being the best mechanism as it solves many problems that primary discovery through non-DNS mechanism introduces, such as complications with multiple CDN hosting.

However, it also makes sense to allow a DoH client to query a DoH server about its properties once it has discovered the server. The method of using HTTPS to confirm designation is provided as an alternative to DNSSEC, which doesn't enjoy universal deployment even today.

Hey Stephane,

Do you have specific concerns with our approach? Is your question about the concept of Adaptive DNS as a whole (zones designating DNS servers for future queries), or about the specific implementation mechanism (using PvDs, where the designation comes from)?

Thanks,
Tommy Jensen
================================================
The latest in Windows Internet Protocols:
  Native gRPC support: https://aka.ms/grpcblogpost
  DNS over HTTPS: https://aka.ms/dohblogpost

________________________________
From: Add <[e-mail]> on behalf of Tim Wicinski <[e-mail]>
Sent: Thursday, July 23, 2020 8:02 AM
To: Stephane Bortzmeyer <[e-mail]>
Cc: Ben Schwartz <[e-mail]>; David Schinazi <[e-mail]>; ADD Mailing list <[e-mail]>; Tommy Pauly <tpauly=[e-mail]>
Subject: [EXTERNAL] Re: [Add] New Version Notification for draft-pauly-add-resolver-discovery-01.txt

As a DNS person, I want to use DNS for that discovery mechanism.
I imagine that non-DNS people want to use non-DNS mechanisms for that discovery.

I would ask how do you see this draft as a better solution than the add-resinfo draft.

tim


On Thu, Jul 23, 2020 at 10:13 AM Stephane Bortzmeyer <[e-mail]<mailto:[e-mail]>> wrote:
On Wed, Jul 22, 2020 at 04:34:41PM -0400,
 Ben Schwartz <[e-mail]<mailto:[e-mail]>> wrote
 a message of 232 lines which said:

> This draft's PvD mechanism (+PUSH) also seems to overlap considerably with
> draft-schinazi-httpbis-doh-preference-hints,

Both proposals (draft-schinazi-httpbis-doh-preference-hints and
draft-pauly-add-resolver-discovery) have in common that a Web server
can tell the user agent which DNS resolver to use for subsequent
queries. I'm not convinced it is a good idea. Before discussing the
details of the proposal, did anyone write a pros-and-cons document
about this idea, which is quite new for the Internet?

--
Add mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/add<https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Fmailman%2Flistinfo%2Fadd&data=02%7C01%7CJensen.Thomas%40microsoft.com%7C070284a795a24cf1b9ce08d82f197fb9%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C637311133847937735&sdata=k8oQTjQvkhq3kqDxq%2FhweUlvJr7yvQ%2BhKAEio3nTxT4%3D&reserved=0>