Re: [TLS] Proposed text for removing renegotiation

[Martin Thomson <[e-mail]>]

On 28 May 2014 03:00, Yoav Nir <[e-mail]> wrote:
> 2. Connections that carry so much traffic and last so very long that you
> really need to rekey. This was discussed at the meeting, partially on Jabber
> ([1]). It’s not that common for the web, but I’m assured that XMPP
> connections sometimes last basically forever, so they need rekeying. My
> proposal there (at 19:40) was for this use case. Martin is right that it
> adds “dead air”, but all these use cases are not that delay sensitive, and I
> think it’s worth it to get the simplified state machine.

I think that we might be able to mitigate the dead air problem.  At
least if the client initiates the action.

My concerns with this approach are with the obvious tension between
the obvious goal and the goals of the application/API.

The establishment of an entirely new context is only truly useful if
it is treated as such.  But the goal of applications in this context
is to continue to operate seamlessly.  Many of the concerns arising
from renegotiation arise from the way that applications are almost
willfully ignorant of the transition in states.

As Martin (R) observes, renegotiation is fairly precisely defined.  If
you strictly respect it, it can be safe...in theory.  It's just that
it's rarely afforded that respect.