I'm investigating the new network framework for use in a server scenario. I can write a trivial server with no security that detects a tls/ssl clientHello and then report that security was needed. Likewise I can write the server code so that it is preconfigured with a certificate so as to use an appropriate level of security. However, I currently can't see how I'd go about configuring the security of the protocol with some sort of bypass/fallback so as to say: if no "clientHello" received then treat comms as *****/insecure and continue? (and yes, I'd later check the tls status of the connection to decide what is appropriate to do...)
Am I missing something in the API, or are my requirements a bit of an outlier?