RE: [SLE] need to make sure data is gone ...
True, I agree that the only way to ensure 0% data recovery is melting the drives . . . . How many people have a scanning electron microscope and would be willing to use it for this purpose? At a computer recycling facility I am familiar with, they use the badblock application and overwrite the entire drive about five to six times with random data. Mark -----Original Message----- From: Mark A. Taff [mailto:marktaff@comcast.net] Sent: Thursday, January 19, 2006 1:29 PM To: suse-linux-e@suse.com Subject: Re: [SLE] need to make sure data is gone ... On Thursday 19 January 2006 10:40, Deep Thinker wrote:
Hello all,
I am in the process of moving and I want to get rid of (sell, donate, give away) all of my old spare parts and self made boxes. I have about 20 hard drives of various sizes and I need to get rid of them as well. My question is, how can I make absolutely sure that all the data has been erased from those drives before I give them away.
The drives all came from clients that I did upgrades for and they did not want to keep the old systems. I have no idea which drives have sensitive information on them and which ones do not.
Thanks for any suggestions in advance.
-- d33p th1nk3r
They only way to make 100% sure the data is unrecoverable is to physically melt the aluminum platters, either in a furnace, or perhaps with an oxy-acetylene torch. The reason for this is that regardless of how to rewrite BS data to the drive, old copies of the data can be recovered using a scanning tunneling microscope from the magnetic molecules at the periphery of each bit. To eliminate this, you have to completely destroy the order of the disk, and that requires changing the physical state of the metal in the disk platters. If you have a mechanic friend, I suggest you grab some beer, and make a party of torching to drives! Mark -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
On Thursday 19 January 2006 13:38, Allyn, Mark A wrote:
True, I agree that the only way to ensure 0% data recovery is melting the drives . . . .
How many people have a scanning electron microscope and would be willing to use it for this purpose?
At a computer recycling facility I am familiar with, they use the badblock application and overwrite the entire drive about five to six times with random data.
It is probably unlikely that someone would value the information enough to justify the expense of that level of recovery. However, people do keep information on their drives that are priceless. For example, what about the lady who takes a naked picture of herself and e-mails it to her boyfriend? What would she pay to make sure that picture is never recovered from a discarded hard drive? How embarrassing if her father found that picture in a google image search? I am certain that if people knew that the data could be recovered, even if it was expensive to do and very unlikely to occur, most of them would choose the just-to-be-on-the-safe-side option of melting their hard drive platters. For myself, I write pseudo-random data to old discs repeatedly so they are somewhat safe for temporary storage. Then periodically I pound them to smithereens with an old California framing hammer I have--very therapeutic--and then torch the platters.
Anyone give you this link yet? http://dban.sourceforge.net/ B-)
On 1/19/06, Allyn, Mark A
True, I agree that the only way to ensure 0% data recovery is melting the drives . . . .
How many people have a scanning electron microscope and would be willing to use it for this purpose?
At a computer recycling facility I am familiar with, they use the badblock application and overwrite the entire drive about five to six times with random data.
Mark
As of a year ago the US DOD calls for 3 overwrites if the data is sensitive but not classified. For classified they perform physical destruction (after all disk drives are cheap). Also, I have not read about the scanning electron microscope working with drives made since 1995 or so. The old ones are so low in density it is possible to read the remnants of the previous bits. With the newer drives I don't think they even record individual bits. Instead they have a base 16 technique they use. Effectively they have 16 magnetic patterns (like hex) they can write to the disk in a given spot. Apparently figuring out which of the 16 patterns was on the disk prior to the current pattern is very difficult even with a electron microscope.
From what I understand even figuring out the current values of the magnetic fields requires some pretty fancy mathematics. The magnetic signal is just plain noisy and it is hard to pull legitimate info out of it. Greg -- Greg Freemyer The Norcross Group Forensics for the 21st Century
On Thursday 19 January 2006 15:02, Greg Freemyer wrote:
Also, I have not read about the scanning electron microscope working with drives made since 1995 or so. The old ones are so low in density it is possible to read the remnants of the previous bits. With the newer drives I don't think they even record individual bits. Instead they have a base 16 technique they use. Effectively they have 16 magnetic patterns (like hex) they can write to the disk in a given spot. Apparently figuring out which of the 16 patterns was on the disk prior to the current pattern is very difficult even with a electron microscope.
That strikes me as dubious Greg. Do you have a reference for that? Storing one of 16 possible values requires a nibble--4 bits (2^4). Imagine this as a 2x2 grid, with each pixel being either light or dark. Even if the disk writes 4 bits at a time(effectively one of 16 values), it is still just 0's and 1's. You would have to have 16 unique possible values being measured for a true hexadecimal system, like 16 levels of voltage. Thanks, Mark
Allyn, Mark A wrote:
True, I agree that the only way to ensure 0% data recovery is melting the drives . . . .
How many people have a scanning electron microscope and would be willing to use it for this purpose?
Are there many on eBay? ;-)
participants (5)
-
Allyn, Mark A -
Brad Bourn -
Greg Freemyer -
James Knott -
Mark A. Taff