/ FEATURE
it of course because the index has no knowledge of the files.
However, using a tool such as WinHex or the Forensic Toolkit
will allow someone who knows how to click `next’ a few times
to scan the hard disk and reconstruct the data. This is the issue
that the researchers have been going on about - what most us-
ers consider to be a properly wiped disk is in fact anything but.
Properly erasing a disk isn’t very difficult and most operat-
ing systems now come with built-in support to do just that.
Now that you know, the why and how of standard data recov-
ery, you are also in a position to know how to stop it. We just
need to make sure that every bit of information on the disk
has been erased. The easiest way to do this is to write zeros
to the disk. We start at the very beginning and keep on going
until the disk is full. In effect, every bit on the disk would read
as zero and the data would effectively be irretrievable. For
example, performing this on a Linux machine (perhaps from a
Linux live CD) can be achieved with this command [3]
cat /dev/zero > /dev/sda
where /dev/zero generates a continuous stream of zeros and
/dev/sda represents the first hard disk. This command will
keep writing zeros until the disk is full.
ProPerly erasing a disk
isn’t very difficult and most
oPerating systems now
come with built-in suPPort
/ The Hidden Flaw: Bit Archaeology
This technique looks foolproof at first sight, but there is a way
around it. It is still largely theoretical, but the idea is that if you
know what data was used to overwrite a given bit, you can, with
the correct hardware, work out what was there previously. This
is because hard disks don’t actually write ones and zeros to
disk but instead leave a magnetic charge that’s close to either to write ones or zeros, an attacker is denied any easy way of
of the two. This isn’t a problem from the user’s point of view, recovering data. However, the DoD recommendation is for seven
as the drive electronics always return either a one or a zero. passes whilst other standards suggest up to thirty-five passes [4].
But if we assume that an investigator has this technology, and Gutmann’s theory has been challenged, however. Gutmann
he knew that we’d simply written zeros, he could look at the insists that the specific patterns he suggests allow for safe
magnetic charge and see how close to zero it actually is. From erasure, as they take various different types of hard disk
this, he could make a very good guess as to whether the bit was physical storage into account. There are many ways to store
previously a one or a zero and thus reconstruct the data. data physically on the disk, especially in the last decade or
To get around this problem, we need to randomise things a so. Gutmann’s patterns were designed to ensure that the data
bit. There are various different specifications for this, the most would be safely erased, regardless of how the disk works
famous by Peter Gutmann [4]. This technique involves several internally. This proposition has been challenged by Feenberg
passes across the disk, each time using a different pattern. [6] at the National Security Agency (NSA), who insists that on
The idea is that if these patterns are followed, several layers modern hardware a simple erase is more than sufficient to
of previous data will be completely random and therefore the ensure that data cannot be recovered.
investigator cannot simply look at the voltages to determine Gutmann is considered an expert, but the NSA is not exactly
the last “real” value that was stored for each bit. low-tech either. The problem is, they can’t both be right –
The DoD has put together its own recommendation, which can either zeroing a disk is secure or it isn’t. At least we can safely
be found in DoD 5220.22 M [5]. Again the idea is that by overwrit- say that zeroing a disk is far better than not doing so. It will
ing the data multiple times, it will become much harder to work prevent the vast majority of people from recovering your data,
out what that data used to be. By randomly choosing whether if they did bothered to look for it. But what if we want to take
16 Digital / ForensicS
DF1_15-19_2nd Feature.indd 16 29/10/09 5:07:33 pm
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52