Christian Huitema <[email protected]> wrote:
>
> The basic QUIC handshake will be 1-RTT before sending the first query,
> with two exceptions:

Thanks for those details!

> Using 0-RTT is a trade-off between security and performance, because
> 0-RTT packets can be subject to replay attacks. That's true for 0-RTT in
> QUIC and also 0-RTT in TLS. If you are really concerned about privacy,
> the prudent decision is to not use 0-RTT.

Correct me if I'm wrong, but my understanding is that the 0RTT replay
attack is not a privacy problem, but is a problem if the payload has
undesirable side-effects. The 0RTT privacy problem is the same as for TLS
session resumption: the session details can be used to track clients.

For privacy-conscious clients, I think it makes sense to use session
resumption for the lifetime of a particular layer 2/3 network connection,
and drop session tokens when roaming to a different connection. So you
benefit from the improved performance while the server has other ways to
track you, but it's harder for the server to track clients from place to
place.

(this is more of a stub -> recursive concern rather than recursive ->
authoritative)

> If 0-RTT is enabled, QUIC performs better than either UDP or TCP in all
> scenarios; if it is not, QUIC still performs slightly better than TCP or
> TLS, because it does not suffer from head of line blocking.

QUIC is something nice to look forward to :-)

Tony.
-- 
f.anthony.n.finch  <[email protected]>  http://dotat.at/
Bailey: West 5 to 7, decreasing 4 for a time. Very rough, becoming rough. Rain
then showers. Good occasionally poor.

_______________________________________________
dns-privacy mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dns-privacy

Reply via email to