On Thu, Sep 12, 2019 at 4:58 AM Petr Špaček <[email protected]> wrote: > > On 12. 09. 19 7:37, Rob Sayre wrote: > > On Wed, Sep 11, 2019 at 2:53 AM Timm Boettger <[email protected] > > <mailto:[email protected]>> wrote: > > > > Hi all, > > > > Rob Sayre has pointed me to this thread. I am an author of the linked > > paper... > > > > He has pointed out some confusing and outdated information in the > > paper, that I would like to clarify... > > > > > > Let me further clarify: I thought these were minor corrections, and I > > didn't want these small flaws to undermine a great point the paper made in > > its Section 3: > > > > "Implementing out-of-order delivery via TLS is akin to (re-)implementing > > the stream multiplexing part of SCTP, QUIC or HTTP/2.0." > > It should be noted that out-of-order query processing is in no way unique to > TLS, it is the same for TCP (see RFC 5966 section 6).
The Google Public DNS DoT implementation supports out-of-order responses so I am surprised that your experiment found that only CloudFlare's implementation supports out-of-order delivery. Maybe this is changed in the updated version of your report. -Puneet > > It should also be noted that Knot Resolver (which I'm working on), BIND and > also modern versions of Unbound _do_ support out-of-order query processing so > the point above it moot. (I'm not sure about PowerDNS but I would be > surprised if it did not have out-of-order processing support as well.) > > Missing support on a particular service is not caused by lack of > implementations in servers, maybe it is a problem "added" by layers on top. > > Petr Špaček @ CZ.NIC > > _______________________________________________ > dns-privacy mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/dns-privacy _______________________________________________ dns-privacy mailing list [email protected] https://www.ietf.org/mailman/listinfo/dns-privacy
