On Fri, Apr 24, 2009 at 5:30 PM, Mike Gill <[email protected]> wrote:
> If the computer boots to a Truecrypt boot loader, and waits for a valid
> password to be supplied so that the disk can be decrypted and Windows may at
> that point come into the picture and begin loading; you're saying this
> defeats all that? It's not a decryption tool.

  Ah.  No, it won't defeat a encryption tool like that.  If you need
to provide a password to decrypt the disk, it won't bypass that.  I
was thinking of disk decryption with a hardware gadget and was
assuming that was present.  Bad assumption on my part.

  This scenario sounds plausible though:

1. Attacker preps system by inserting his bootable debugger; leaves vicinity.
2. Users comes in, boots system, enters decrypt key, leaves vicinity
with system running at an OS password prompt (maybe screen lock).
3. Attacker comes back, bypasses logon/screen lock with debugger.

  It is, of course, a best practice to never leave the system
unattended while the decrypt is active.  But many take the stance that
"Once booted, the system is safe from hard disk analysis; the only way
they can read the hard disk directly is to shut it down and remove it;
that will clear the decrypt key."

  Whether the scenario I describe above is really a use of this new
attack, or just a failure to follow best practices for disk
encryption, is left to the reader.  :)

-- Ben

~ Finally, powerful endpoint security that ISN'T a resource hog! ~
~ <http://www.sunbeltsoftware.com/Business/VIPRE-Enterprise/>  ~

Reply via email to