On Wed, Mar 04, 2020 at 05:09:59PM -0800, Watson Ladd wrote:
> > Because it MUST be possible for clients and servers where only
> > one side supports reuse to *interoperate*. That's the whole
> > point of IETF standards.
>
> A failure to resume does not break the connection. Tickets may age out
> anyway, or the server might have dropped state on restart, etc. So
> there is no interoperability problem.
That's sophistry. The two sides interpret the signal in incompatible
ways, leading to the client not getting the fresh ticket it needs to
continue to communicate efficiently with the server.
Especially with the much larger PQ certificates coming soon, resumption
becomes increasingly important.
Performance-impacting breakdown in signalling due to incompatible
interpretation of specifications is loss of interoperability.
--
Viktor.
_______________________________________________
TLS mailing list
TLS@ietf.org
https://www.ietf.org/mailman/listinfo/tls
