Re: [dns-privacy] Study on the adoption and performance of DNS over QUIC

2022-02-10 Thread Christian Huitema
On 2/10/2022 12:34 PM, James Cloos wrote: "CH" == Christian Huitema writes: CH> and you get 40% of names served by a small number of servers. For that set, CH> I would expect that the typical DoQ query will be a session resumption. does quic session resumption correctly work in the face of

Re: [dns-privacy] Study on the adoption and performance of DNS over QUIC

2022-02-10 Thread James Cloos
> "CH" == Christian Huitema writes: CH> and you get 40% of names served by a small number of servers. For that set, CH> I would expect that the typical DoQ query will be a session resumption. does quic session resumption correctly work in the face of any-cast servers? IU that tls s-r

Re: [dns-privacy] Study on the adoption and performance of DNS over QUIC

2022-02-10 Thread Christian Huitema
I am not sure. There is a fair bit of concentration in the "authoritative server" market. I see measurements (https://ithi.research.icann.org/graph-m9.html) showing that Go Daddy, Cloudflare and AWS have a fairly big share of this market. Add Akamai, Dyn and Google to the mix, and probably a

Re: [dns-privacy] please adopt draft-dkgjsal-dprive-unilateral-probing as a WG work item Re: New Version Notification for draft-dkgjsal-dprive-unilateral-probing-02.txt

2022-02-10 Thread Peter van Dijk
On Thu, 2022-02-03 at 12:59 -0500, Daniel Kahn Gillmor wrote: > Hi Peter, DPRIVE folks-- > > On Thu 2022-02-03 11:03:35 +0100, Peter van Dijk wrote: > > Speaking only for myself: some of the parts still seem too prescriptive > > to me (but I know I haven't been clear on what parts!). Examples:

Re: [dns-privacy] Study on the adoption and performance of DNS over QUIC

2022-02-10 Thread libor.peltan
Anyway, do you think that the "typical" recursive-to-authoritative query in DoQ era will be a session resumption, or a clean new connection without any chance of 0-RTT or counter-amplification-limit-token? I agree that both cases could be explored, but the latter one is still a good start.

Re: [dns-privacy] Study on the adoption and performance of DNS over QUIC

2022-02-10 Thread Mike Kosek
> Mike Kosec mentioned on Twitter that "We do not use TLS session resumption." > (https://twitter.com/MikeKosek/status/1491351241695592454 > ). This are a very > serious limitations of the test client. The performance of DNS over QUIC >