The /dev/random device produces fairly strong random numbers. In order to do that, it
needs to get fairly random input. It gets this input by sampling keypress timings,
mouse movements, the like. So what happens when you dd it to /dev/hda is it writes
out a small amount of random data, based on it's current 'entropy pool' then waits for
you to wiggle your mouse and press keys to re-fill the
'entropy pool' before writing out more data. Gets you fairly strong randomness, but
takes a loong time to generate a whole hd full.
The /dev/urandom device (u=symbol for micro) will continue to generate "random" output
after the entropy pool is empty. This results in less randomness, but can generate
vast quantities of pseudo-random data.
/dev/zero is even faster, but far less random;)
Anyway, you stated that you didn't need it to be a really secure wipe, so it probably
doesn't matter, but IIRC it is easier to recover data overwritten by a consistent
pattern than a random mix. I think more different-pattern wipes are harder to read
through than one pass of any amount of randomness, though.
"Barry L. Kline" wrote:
>
> "Barry L. Kline" wrote:
> >
>
> Thanks everyone for your suggestions! I'm going to: dd if=/dev/zero
> of=/dev/hda
>
> Barry
>
> _______________________________________________
> Redhat-list mailing list
> [EMAIL PROTECTED]
> https://listman.redhat.com/mailman/listinfo/redhat-list
--
-------------------------------------
Sam Bayne - System Administrator
North Seattle Community College
[EMAIL PROTECTED] (206)527-3762
=====================================
_______________________________________________
Redhat-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-list
