Re: [Doh] [Ext] Re: Associating a DoH server with a resolver

Adam Roach <> Wed, 24 October 2018 03:40 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 15486130DDA for <>; Tue, 23 Oct 2018 20:40:21 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.88
X-Spam-Status: No, score=-1.88 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, T_SPF_HELO_PERMERROR=0.01, T_SPF_PERMERROR=0.01] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id vhUThWYThMKS for <>; Tue, 23 Oct 2018 20:40:17 -0700 (PDT)
Received: from ( [IPv6:2001:470:d:1130::1]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by (Postfix) with ESMTPS id 46C06130DE5 for <>; Tue, 23 Oct 2018 20:40:15 -0700 (PDT)
Received: from ( []) (authenticated bits=0) by (8.15.2/8.15.2) with ESMTPSA id w9O3dp9n095790 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128 verify=NO); Tue, 23 Oct 2018 22:39:52 -0500 (CDT) (envelope-from
X-Authentication-Warning: Host [] claimed to be
To: Eric Rescorla <>, Martin Thomson <>
Cc: Paul Hoffman <>, DoH WG <>
References: <> <> <> <> <>
From: Adam Roach <>
Message-ID: <>
Date: Tue, 23 Oct 2018 22:39:45 -0500
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.13; rv:60.0) Gecko/20100101 Thunderbird/60.2.1
MIME-Version: 1.0
In-Reply-To: <>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Archived-At: <>
Subject: Re: [Doh] [Ext] Re: Associating a DoH server with a resolver
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: DNS Over HTTPS <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 24 Oct 2018 03:40:21 -0000

On 10/23/18 10:18 PM, Eric Rescorla wrote:
> Several points here:
> 1. As a matter of aesthetics, I agree with Martin that domain names 
> would be better.
> 2. Martin sent a link to a method for resolving TXT records on 
> Windows. MacOS has its own API: 
> So, this doesn't seem prohibitive to me.
> 3. It seems like in the use case for which this draft is specified, 
> the whole thing is pretty opportunistic, so IP address certs wouldn't 
> be required.
> 4. There are other uses cases for which it might be nice to have real 
> domain names, in which case the IP address cert thing is a pain.
> For these reasons, I think a domain name in TXT or the like would be 
> better.

If we are going for non-A/AAAA record types, then I would suggest that 
we use the record to directly indicate the URI instead of having to 
synthesize one using a .well-known path. I would also point out that we 
already have a resource record type for converting desired services into 
URIs (RFC 7553).

So, e.g., interested clients would query for a URI record of type, and get back a full HTTPS URL as a response.