I have the following question about TLS security:

Assume TLS-PSK protected HTTP with AES256-CBC cipher. When a TLS connection is established, client sends some encrypted data, where the plaintext is well known to an attacker.
For example, it can be a HTTP request, like:

GET /some_resource HTTP/1.1

Can an attacker use intercepted ciphertext and known plaintext it to reveal the secret key?

Does TLS have any features to protect from that?

  • 1
    $\begingroup$ Does TLS have any features to protect from that Yes... AES! $\endgroup$ – rath Apr 26 '14 at 16:16

This question boils down to, "Are there known-plaintext attacks against AES-256-CBC?" The answer is: No, no such attacks better than brute force are known. This would constitute a catastrophic break in AES-256-CBC, and any cipher broken in such a way would be abandoned rapidly.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.