[445] in Kerberos_V5_Development
Re: Kerberized Telnet (A warning)
daemon@ATHENA.MIT.EDU (Jon A. Rochlis)
Sat Nov 17 18:29:25 1990
From: jon@MIT.EDU (Jon A. Rochlis)
To: Paul Borman <prb@krystal.cray.com>
Cc: bcn@cs.washington.edu, krbdev@ATHENA.MIT.EDU
In-Reply-To: Your message of Sat, 17 Nov 90 15:42:20 -0600.
Date: Sat, 17 Nov 90 18:28:51 EST
Well, I would guess that if an attacker knows EXACTLY what you are sending
when in your data stream, it wasn't much use to do the encryption in the
first place.
Wrong! I may want integrity more than I want privacy. Encryption is
a technique used to achive both integrity and privacy, not just
privacy.
From a brief look at the code it looks like you are using a key
autokey (KAK) cipher. The key stream is only a function of the
initial vector, not of the cleartext (or ciphertext). This means you
immediately resync after a ciphertext modification. On the other
hand, if you use a stream cipher that uses a key stream which is a
function not only of the initial vector but also the cipher (or clear)
text, ciphertext errors will propagate (and thus be much more likely
to be detected). This may be significant.
-- Jon
