RE: Peter Gutmann data deletion theaory?

[Jeremy Epstein <jeremy.epstein () webmethods com>]

Like anything in security, "it depends".  In particular, it depends on what
the assumed adversary motivations and capabilities are.  If the adversary is
a nation-state with electron microscopes and other expensive devices, and
the disk is believed to have held highly classified information, it's
clearly true that the only way to destroy the data is to burn the disk (and
in the right way).  If, on the other hand, the adversary is someone who's
randomly buying used computers in hopes of finding carelessly deleted files,
then a full reformat is quite enough to cause them to move on to the next
machine - they're not going to have the motivation or equipment to delve
into a randomly selected disk.

Where in between these two extremes it's necessary to burn the disk is an
exercise left to the reader ;-)  You really have to do a risk analysis... If
it's cheaper / easier / less dangerous for the adversary to dumpster dive to
get hardcopies or bribe someone or hack into the system, then destroying the
hardware is putting the effort in the wrong place.  For a lot of classified
systems, the assumption is that obtaining used disks is a low cost attack,
so it's cost effective to use destruction.

--Jeremy

> -----Original Message-----
> From: Jared Johnson [mailto:jaredsjazz () Yahoo com] 
> Sent: Wednesday, July 20, 2005 7:49 PM
> To: focus-ms () securityfocus com
> Cc: bugtraq () securityfocus com
> Subject: Peter Gutmann data deletion theaory?
>
> All,
>
> Do you all agree with Peter Gutman's conclusion on his theory 
> that data can never really be erased, as noted in his quote below:
>
> "Data overwritten once or twice may be recovered by 
> subtracting what is expected to be read from a storage 
> location from what is actually read. Data which is 
> overwritten an arbitrarily large number of times can still be 
> recovered provided that the new data isn't written to the 
> same location as the original data (for magnetic media), or 
> that the recovery attempt is carried out fairly soon after 
> the new data was written (for RAM). For this reason it is 
> effectively impossible to sanitise storage locations by 
> simple overwriting them, no matter how many overwrite passes 
> are made or what data patterns are written. However by using 
> the relatively simple methods presented in this paper the 
> task of an attacker can be made significantly more difficult, 
> if not prohibitively expensive."
>
> It seems that the perhaps the only real way to rid your Hard 
> Drives of data is to burn them. 
>
> I'd love to hear some thoughts on this from security and data 
> experts out there.