Re: [TLS] Last Call: draft-hoffman-tls-additional-random-ext (Additional Random

[Dean Anderson <[e-mail]>]

On Mon, 26 Apr 2010, Marsh Ray wrote:
> http://tools.ietf.org/html/rfc2246 :
> > 7.4.1.2. Client hello
> > 
> > [...]
> >
> >    random_bytes
> >        28 bytes generated by a secure random number generator.
> 
> Not pseudorandom, "generated by a secure random number generator".

Unfortunately, my CRC handbook is packed up at the moment. I think 
this is right: From
http://www.sunny-beach.net/random_numbers/manual/173.htm

   1. The random numbers should pass statistical tests of randomness.
   2. It should be difficult to predict the output of the random number 
generator from observing some previous outputs of the random number 
generator.
   3. It should be difficult to create the same operating conditions as 
the random generator and then duplicate previously produced output.


Most implementations use a PRNG with hopefully these properties.  I just
read an article on phase-space analysis that shows that many PRNG's that
pass (1) above actually fail at (2) and (3).

Most people don't have access to truly random numbers.  PC's and most
computers don't even have the capability to even get truly random
numbers without some kind of extra hardware.

So where are these non-pseudo, truly secure random numbers going to come
from? I think Nowhere, they don't usually exist in practice, except
maybe at the NSA.

		--Dean

-- 
Av8 Internet   Prepared to pay a premium for better service?
www.av8.net         faster, more reliable, better service
617 256 5494