OT Former IRS official Lerner’s hard drive crash?

[John Doe]

Now that she's been quoted for saying some nasty things about
conservatives while working for the IRS.

 

[John Doe]

The weird thing about this story is that some news reports (NBC)
say the hard drive was "scratched". And they aren't sure whether
the scratch was put there intentionally or not.

Since when can you scratch a hard drive?

 

[PAS]

The weird thing about this story is that some news reports (NBC)
say the hard drive was "scratched". And they aren't sure whether
the scratch was put there intentionally or not.

Since when can you scratch a hard drive?

Since you realize you could be facing criminal charges.

 

[John Doe]

"John Doe" <always.look message.header> wrote

Since you realize you could be facing criminal charges.

You are a regular here? How do you scratch a hard drive?

 

[Paul]

The weird thing about this story is that some news reports (NBC)
say the hard drive was "scratched". And they aren't sure whether
the scratch was put there intentionally or not.

Since when can you scratch a hard drive?

You can gouge one.

This is my first failed hard drive.

http://i61.tinypic.com/16hu5p4.jpg

That drive made a neat sound effect, when the gouging happened.
It sounded like a windup clock spring had sprung. There was
a "sproing" noise. What happened was, the head lock solenoid
jammed, the head assembly tried to move anyway, and the head
assembly got pressed or twisted into the platter while it was
spinning.

To put a scratch radially on a platter, all it would
take is the head on the end of the arm, detaching during
a seek. But then, you would only expect one surface of the
N surfaces in the drive to be scratched.

And you can work on platters with this, if they're no longer
suited to rotational data recovery. This remains a "theoretical"
attack, as I've never read of any practical examples (like
this thing being used to bring evidence in a court case).
There are supposed to be more than a thousand of these
in existence, and so you might find one in a university
lab (metallurgy department maybe). It's unclear whether
the sampling area on one of these, would be big enough
to fit an entire platter. They could certainly have an
X-Y gross positioner on the sample table - but does this need
vacuum while it works ?

http://en.wikipedia.org/wiki/Magnetic_force_microscopy

The sticker on the drive, usually covers a screw or two.
The disk drive companies do that, for detection of
tampering (user opens drive, closes drive up, then
expects warranty support). On the above pictured drive,
the action of removing that sticker, left stretch
marks in the plastic. It was easy to see I had tampered
with it. The adhesive on that particular label was pretty
good. The label did not tear while being removed - the
plastic was good stuff. But the plastic also ended up
with stress marks in it. Black plastic with white stress
marks.

Someone at the factory could provide a fresh sticker.
You would need a pretty extensive collection of
anti-tampering stickers, to have the exact one
needed for that drive, then adhere it to the drive.

And there is an easier way to gain access. The silver-colored
stickers cover things like servo writer ports. You don't
have to undo any screws, if you just want some means
to get at a platter. The silver stickers could also
be a bit easier to replace. There is no printing on them.

Only certain scratch patterns are "feasible" if put
there by a hardware failure. Concentric rings like
my example, make perfect sense. An arc-shaped scratch
from head loss during seek, is a possibility. Not
any random scratch makes sense, so whether the
"drive did it", the scratch pattern would give this
away. The scratch needs the right trajectory.

Even during head unload, the platter is still spinning.
The motor doesn't stop, until the head assembly is run
up the landing ramp (on a modern drive). So scratch
marks should bear some relationship to "spinning".
The heads also will not load, unless the motor controller
signals it is up to speed. The platter spins, in order
for the "flying" effect to exist. Not spinning fast
enough, could cause a head crash. On power failure, the
heads must be retracted - in a hurry. On the ancient
ramp-less drives, the heads were parked in the landing
area, and would do so on power fail as well.

So an expert would have plenty of little things to
feed into determining the real failure mode. If
you're going to fake a scratch, you have to know
your physics to get away with it.

Paul

 

[Loren Pechtel]

The weird thing about this story is that some news reports (NBC)
say the hard drive was "scratched". And they aren't sure whether
the scratch was put there intentionally or not.

Since when can you scratch a hard drive?

Head failure.