Re: [keyassure] Bare keys again

[Paul Wouters <[e-mail]>]

On Wed, 23 Mar 2011, Douglas Otis wrote:

>> http://blog.torproject.org/blog/detecting-certificate-authority-compromises-and-web-browser-collusion 
>> 
>> This is another clear signal that DANE is needed, and the need for the
>> X509 storage format is unneccessary. So people can simply take control
>> of SSL for servers within their own DNS zones.
> Could this end the practice of stapling Certs to server responses that might 
> be cached for weeks to support ultra-fast browsers?  Perfect MitM.  Do these 
> browser insist on seeing valid nonce extensions, as this would hurt 
> performance?

I was also pretty disappointed to see that when you send a
trusted_ca_keys: pre_agreed value, that this is totally ignored on
servers, including openssl.org. I have a rough patch for openssl I used
for testing, but not one that takes any of the other RFC 6066 options (yet)

In fact, if I understood the versioning right, basically no one is
running TLS v1.2 yet, and the comments in the openssl 1.0.1 snapshot
suggested it had only partially started on TLs v1.1.

I did not expect this backlog on the TLS implementations.

Paul