Recycling old pc, erasing hard drive

Status
Not open for further replies.
Originally Posted By: XS650
Originally Posted By: Drew99GT
Even if the platter is smashed into tiny little pieces?
15.gif
13.gif



Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).
 
Originally Posted By: PandaBear
Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).


Do you know how they destroy drives at Livermore Labs?
wink.gif
 
Darik's boot & nuke for the win! Have an old laptop I'm putting on ebay. I do not like buying "mostly complete" machines minus hard drives as the seperate shipping and hassles for seperate, appropriate old hard drives adds to the price, and I would not do that to someone else.

Ran this painless app this morning and have the highest confidence, it goes through at least three seperate processes of wiping.
 
Originally Posted By: moribundman
Originally Posted By: PandaBear
Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).


Do you know how they destroy drives at Livermore Labs?
wink.gif



They take all the PCB out and burn it to ash, take the magnetic media and sand blast the surface until all the magnetic coating is gone, then melt them along with all metal parts in a smelter. In the end nothing is left but burn/melted raw material.
 
Originally Posted By: moribundman
Originally Posted By: PandaBear
Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).


Do you know how they destroy drives at Livermore Labs?
wink.gif



Now some poor soul at the NSA gets to monitor BITOG. Poor guy (or girl), he or she will be ruined.
 
Originally Posted By: moribundman
Originally Posted By: PandaBear
Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).


Do you know how they destroy drives at Livermore Labs?
wink.gif



Don't they let former Chinese nationals take them home :)

Quote:

The laboratory has attracted negative publicity from a number of events. In 1999, Los Alamos scientist Wen Ho Lee was accused of 59 counts of mishandling classified information by downloading nuclear secrets—"weapons codes" used for computer simulations of nuclear weapons tests—to data tapes and removing them from the lab. After ten months in jail, Lee pled guilty to a single count and the other 58 were dismissed with an apology from U.S. District Judge James Parker for his incarceration.[1] Lee was suspected for a time of having shared U.S. nuclear secrets with China, but investigations found this not to be true.[2][dubious – discuss] In 2000, two computer hard drives containing classified data were announced to have gone missing from a secure area within the laboratory, but were later found behind a photocopier; in 2003, the laboratory's director John Browne, and deputy director, resigned following accusations that they had improperly dismissed two whistleblowers who had alleged widespread theft at the lab. The year 2000 brought additional hardship for the laboratory in the form of the Cerro Grande Fire, a severe forest fire that destroyed several buildings (and employees' homes) and forced the laboratory to close for several days.

In July 2004, an inventory of classified weapons data revealed that four hard disk drives were missing: two of the drives were subsequently found to have been improperly moved to a different building, but another two remained unaccounted for. In response, director Peter Nanos shut down large parts of the laboratory and publicly rebuked scientists working there for a lax attitude to security procedures. In the laboratory's August 2004 newsletter he wrote, "This willful flouting of the rules must stop, and I don't care how many people I have to fire to make it stop". Nanos is also quoted as saying, "If I have to restart the laboratory with 10 people, I will". However, a report released in January 2005 found that the drives were in fact an artifact of an inconsistent inventory system: the report concludes that 12 barcodes were issued to a group of disk drives that needed only 10, but the two surplus barcodes nevertheless appeared on a master list. Thus, auditors wrongly concluded that two disks were missing. The report states, "The allegedly missing disks never existed and no compromise of classified material has occurred". This incident is widely reported as contributing to continuing distrust of management at the lab. In May 2005, Nanos stepped down as director.


Quoted from http://en.wikipedia.org/wiki/Los_Alamos_National_Laboratory

And yes, I realize the quote is about Los Alamos, they probably have similar policies, LOL.
 
If you're trying to deter all but a recovery lab, just pull the PCB off the bottom before tossing it. That's more than enough to discourage even hard-core identity thieves. Takes all of thirty seconds.

The PCBs and related firmware on a HDD are custom calibrated to the actual platters in that drive. Even installing an exact replacement PCB (a tall order once the original is missing) will not be enough to read the media. Specialized equipment will still be required to recalibrate the drive electronics to the platters.

It's a big dollar proposition to recover data once the original PCB is gone or malfunctions. Been there, done that.

If you must be absolutely sure that **** Cheney can't read it, then open the drive and smash up the platters.

I'm not a fan of software-based HDD erasing solutions. I've seen other software that can still reach what was thought "overwritten".
 
Originally Posted By: wavinwayne
Originally Posted By: moribundman
Originally Posted By: PandaBear
Yes for highly sensitive data that is worth recreating (i.e. defense related info, national secret, atomic bomb blue print, nuclear test simulation data).


Do you know how they destroy drives at Livermore Labs?
wink.gif



Now some poor soul at the NSA gets to monitor BITOG. Poor guy (or girl), he or she will be ruined.


Yes, someone here will be watched for sure...
LOL.gif
 
Actually it is quite easy to replace the PCB and recover all the data on the HD after the PCB was taken. It is a bit more complex than just same part numbers of PCB, but the same firmware on the flash memory or ROM on the PCB and having it match the one in the HD's platter. Quite easy if you do recovery for a living since you usually have access to a library / junkyard worth of HD parts.



Regarding to Wen Ho Lee. I used to work with a co-worker whose husband use to work with Lee back in the days. That co-worker's husband said Lee isn't really a spy, but a weirdo/nerd/geek that have some temper and anger management problem, and no body likes him. He also said it is very common to misplace HD and people (not just Lee) took them home all the time, so probably there aren't even a spy or the spy is someone else and scapegoat Lee for it.
 
Originally Posted By: Volvohead
If you're trying to deter all but a recovery lab, just pull the PCB off the bottom before tossing it. That's more than enough to discourage even hard-core identity thieves. Takes all of thirty seconds.

The PCBs and related firmware on a HDD are custom calibrated to the actual platters in that drive. Even installing an exact replacement PCB (a tall order once the original is missing) will not be enough to read the media.


I am sorry, but you are incorrect. I have done this swap numerous times to recover data from drives with baffed PCB's, and even had successful platter-swap recoveries using identical drives.
 
Like I stated earlier. Perform a quick swirel with a 40 amp plasma cutter on the drive itself. OK IT folks tell me that you could reconstruct something that has been exposed to plasma cutting...that would be the day!!
 
We will have to disagree.

On most drives manufactured after 2002, as more adaptive firmware systems that store variables in memory have been developed, simple PCB swaps have become increasingly difficult in the field. Don't even bother if you ever touch the head stacks.

As of 2008, you really need a lab inventory with the latest drives, and even the labs often need the original PCB to recover stored parameters. If you pulled it off on a newer drive, then you should play the lottery tonight.

Easy general reading:
http://www.essdatarecovery.com/parts_exchange.asp

For the more technical nuances:
http://www.actionfront.com/whitepaper/Drive-Independent Data Recovery Ver14Alrs.pdf
http://www.actionfront.com/whitepaper/Dr...%20Preprint.pdf

We got an expensive crash course in all of this not that long ago . . . pun intended.
 
Originally Posted By: PandaBear
Quite easy if you do recovery for a living since you usually have access to a library / junkyard worth of HD parts.


I think that was my premise.

Unless it's going to a recovery lab that has a large FW inventory, the average field tech (and most garden variety ID thieves) can't pull it off. And even the labs often need the original PCB to pull the flash parameters for the subject drive.
 
To clarify, FW parameters are still written to the drive platters. But there is increasing use of flash parameter storage on the PCB itself. This is independent from the PCB HW version.

What I understand the labs to do is either read the flash from the old PCB and load it on the donor, or do a chip transplant onto the donor board.
 
Originally Posted By: Volvohead
To clarify, FW parameters are still written to the drive platters. But there is increasing use of flash parameter storage on the PCB itself. This is independent from the PCB HW version.

What I understand the labs to do is either read the flash from the old PCB and load it on the donor, or do a chip transplant onto the donor board.



Parameters and partial firmware have long been written to platters, and the rest to ROM on the PCB or a small piece of flash memory on the PCB. HD from the same model and capacity, sold to the same customer (OEM or retail, which OEM, etc) around the same manufacturing date (+/- a couple of months or less) usually use the same PCB with the same firmware (again, just check the stainless cover). Someone that you are scraping the HD to should be able to find out what you are using as PCB just by looking at the manufacturing date and model/serial/FW version, etc on the stainless steel cover, then go on ebay and search for the exact one (or call up a lab that keep an library of drives) to get the data back. It is easier done than you think.

Platter specific info that are written during self test or manufacturing calibration is not written to PCB, because those info are too large. PCB's parameters are usually only sufficient to boot the drive to firmware stored on the platters (i.e. channel parameters, frequencies, fly height, inertia, PIDs, etc), at least that's how Maxtor's drives work when I do firmware over there.

Now if you are only worried about ID thieves, just blank the drive with zeros. Then they have to use an electron microscope and some HD knowledge to recreate all the bits via residue magnetic signals from before the blank/zero/nuke of the data. This would be very calculation and labor intensive that is only worth doing for national secret and military info.
 
I really don't think we're disagreeing about much here.

If all the OP is trying to accomplish is keeping the neighbors from dumpster diving his drive and viewing little Johnny's christening pix, pulling the drive electronics is usually enough. It takes either a lab at a couple grand or an extremely determined and lucky tech who can find the exact FW and PCB rev. and get it to boot, without destroying it. Manufacturers change those parameters on almost a weekly basis anymore. Nobody's going to go through all that effort and expense over a garden variety household PC drive sitting in the garbage. The professional ID thieves loot it electronically while it's spinning in the guy's office. Now if someone KNEW the secrets to the Iranian nuclear program were on it, that's another story.

If he wants to keep certain secrets forever secret from everyone, then he's going to have to open it up and physically destroy the media. As we both know, even writing a string of zeros isn't enough.

The newest monster density drives have gotten so cranky, it's hard enough to keep them from destroying themselves just from running them.
 
Originally Posted By: Volvohead
We will have to disagree.

On most drives manufactured after 2002, as more adaptive firmware systems that store variables in memory have been developed, simple PCB swaps have become increasingly difficult in the field. Don't even bother if you ever touch the head stacks.

As of 2008, you really need a lab inventory with the latest drives, and even the labs often need the original PCB to recover stored parameters. If you pulled it off on a newer drive, then you should play the lottery tonight.

Easy general reading:
http://www.essdatarecovery.com/parts_exchange.asp

For the more technical nuances:
http://www.actionfront.com/whitepaper/Drive-Independent Data Recovery Ver14Alrs.pdf
http://www.actionfront.com/whitepaper/Dr...%20Preprint.pdf

We got an expensive crash course in all of this not that long ago . . . pun intended.


I think it's very dependant upon the brand and family of drive. These have all been consumer-level drives. Swap candidate has been an identical drive with the exact same model number and physically identical PCB. I have yet to have an incident where recovery was necessary on an enterprise-grade drive..... A lot more redundancy in place with that kind of equipment... People at home... yeah, not so much
wink.gif


The most common drive was the old 40GB and 20GB Fujitsu IDE drives. These would have been pre-2002.

Most recent has been a Maxtor 160GB drive. Post-2002.
 
Status
Not open for further replies.
Back
Top