Re: [Doh] A question of trust (was Re: Draft -09 and WGLC #2)

[Mateusz Jończyk <[e-mail]>]

Hello,

>+A DNS API client MUST NOT use a different URI simply because it was discovered
>+outside of the client's configuration, or because a DNS API server offers an
>+unsolicited response
>+that appears to be a valid answer to a DNS query.

The second part of this sentence is unclear to me. Some people may read it as
prohibiting all Server Pushed messages.
I would reword it as:
	... or simply because some web server offers an unsolicited response
	that appears to be a valid answer to a DNS query.


I think the previous wording in the "Security considerations" section was much
better:
	A client MUST NOT use arbitrary DNS API servers.
	Instead, a client MUST only use DNS
	API servers specified using mechanisms such as explicit configuration.

It only required a rewording of the section "Selection of DNS API server" to
match it.

------------------

By the way, we should clarify the section "Server push". I propose it to read so:

	A DNS API client MUST ignore pushed DNS API requests (see {{RFC7540}}
	Section 8.2) whose pushed request URI is not one that the client
	would have directed the same query to if the client had initiated the
	request.

(which is a modification of the text proposed in
	https://github.com/dohwg/draft-ietf-doh-dns-over-https/pull/185
and suits the comments there).



Greetings,
Mateusz Jończyk



W dniu 30.05.2018 o 04:14, Patrick McManus pisze:
> I've proposed https://github.com/dohwg/draft-ietf-doh-dns-over-https/pull/189
> 
> Mateusz, it also harmonizes with the security considerations a bit.