RE: MD5 Collisions and Evidence Integrity

Apparently everyone might want to review:

http://eprint.iacr.org/2004/199.pdf This is pure math so get lots of coffee.

For this to occur, and I will use EnCase as an example because it uses the
Acquisition Hash and then immediate Verification Hash.

You have hash (output) "Z" which is created from Hard-drive "A" in its
current state we will call it (input) "X".

You have found a numerical composition (input) "Y" that will also create
"Z". 

So we have Input "X" Hash= "Z" and Input "Y" Hash= "Z"

You would have to be able to alter the Hard-drive EXACTLY to the point to
make it equal "Y" so that when you hashed output "Z". 

Now lets think about this, EnCase Hashes the drives, acquires the evidence
file(s), then does a verification hash. You would have to do the above in
the middle of that process.

I am not saying it is impossible, but I beat the odds are about the same as
finding two people with the same fingerprints.


______________________________________
Dave Kleiman, CISSP, CISM, CIFI, MCSE
www.SecurityBreachResponse.com




> > Actually MD5 hasn't been broken... yet. A close cousin to MD5 was broken,
not the actual MD5 as we know it. It's still safe to use MD5 for the time
being. >> However I would be on the look out for a replacement if and when
one becomes available. Sure there's SHA1, but of course a pair of digests to
compliment >>each other is prefered.
> > -dosman




-----------------------------------------------------------------
This list is provided by the SecurityFocus ARIS analyzer service.
For more information on this free incident handling, management 
and tracking system please see: http://aris.securityfocus.com