[NNTP] TLS draft question

[Ned Freed]

> Ken Murchison <ken at oceana.com> writes:
> > Eric Rescorla wrote:
> >> I think Ned's right here, if for no other reason than that SSL
> >> implementations can buffer and so you've now got to figure out how to
> >> unbuffer the data. Sorry for not catching that when Id id my
> >> review..
> >
> > No apologies nescessary, but I'm looking for guidance.  What you're
> > both describing seems to be an implementation problem, not a design
> > flaw in the protocol.  What do you suggest we do with NNTP that would
> > differ from the other messaging protocols?

> I'm not sure it's different. My memory was that noone actually tried
> to resume after failed TLS negotiation and that if the protocol
> specs suggested it they were widely ignored. :)

This corresponds exactly to my experience.

> Have I been away too long? :)

For that matter, it has been a couple of years since I tried this. Things
may have changed since then. OTOH, old implementations never seem to
die out completely.

				Ned