Re: Crypto Secure Random Numbers

[Andrew Basterfield <list@lostgeneration.freeserve.co.uk>]

On 07 Feb 2002 04:05:00 +0100
Artur Grabowski <art@blahonga.org> wrote:
> > Theory of operation:
> >   When gettimeofday(2) is called, the microseconds value returned is
unknown> > and unpredictable. It is therefore a source of entropy. By
introducing a

> This time there is not even a need to explain the theory of why your new
> randomness collection method sucks. Simple tests show that it's
completly> useless.

If that random function was called at fixed time intervals (eg every
microsecond) the output would not be at all random. The time imterval does
not have to be exactly 1 microsecond - any fixed interval would create a
non-random interference pattern in the output data as the frequency of the
unix clock 'beats' with the frequency that routine is called (sorry I'm an
elec eng it's the way I visualise it). This is less noticable on a system
with variable load & hence variable latency but consider when the SSH keys
are generated at 1st boot and there are no other processes active?

Leave that stuff to the experts (or learn enough to become one yourself).

A. Basterfield