Re: [TLS] 0-RTT and Anti-Replay

[Colm MacCárthaigh <[e-mail]>]

On Mon, Mar 23, 2015 at 6:12 AM, Martin Thomson
<[e-mail]> wrote:
> Well, Adam described at the interim why this isn't a concern: as the
> "strike register" takes a higher density of entries, the window of
> time over which it applies can be narrowed to limit the potential for
> state exhaustion.  That simply increases the likelihood of a
> rejection, not the attack.

I haven't been able to get my head around this. Doesn't that mean that
an attacker can disable 0RTT for third parties at will? it also means
that the session itself must be invalidated on a shorter duration, or
else we could just collect 0RTT data from before the time window and
replay those.

-- 
Colm