Confidential Data Not Safe On Solid State Disks

Nuke it from orbit by ColdWetDog · 2011-02-17 08:15 · Score: 2, Funny

It's the only way to be sure.

--
Faster! Faster! Faster would be better!

Re:Nuke it from orbit by Anonymous Coward · 2011-02-17 08:25 · Score: 1

I use a TrueCrypt volume inside a TrueCrypt volume. It offers enough protection for my private data and is still more than fast enough for my needs.
Re:Nuke it from orbit by DigiShaman · 2011-02-17 08:41 · Score: 3, Funny

Or in a microwave. That seems to destroy the gates on the chip. 10 seconds on High should be enough. Just be sure to only place the PCB and not the entire drive as they can contain lots metal.

--
Life is not for the lazy.
Re:Nuke it from orbit by XiaoMing · 2011-02-17 08:42 · Score: 1

Careful what you wish for...
http://science.slashdot.org/story/11/02/17/1828232/Sysbrain-Lets-Satellites-Think-For-Themselves
Re:Nuke it from orbit by rossdee · 2011-02-17 08:50 · Score: 1

That may be a little expensive, what about nuking it in a microwave oven?
Re:Nuke it from orbit by MachDelta · 2011-02-17 08:59 · Score: 3, Interesting

The most fun I ever had disposing of a HDD was when I worked as a mechanic. One of the POS systems was being replaced and the drive in it was going to be shredded. It was a slow day then, so I bugged our IT guy to let me have a crack at it. With an evil grin, I took it out to a workbench, stuffed it in a vice, and beat the piss out of the casing with a hammer. Once it was suitably mangled I started taking it apart with a prybar and screwdriver (gotta save those magnets!) until all I had left was the stack of platters. I took them to the 10 ton press in the back and squished it into a platter-pizza. Then I went to the corner and took the Oxyacetylene torch to that sum'bitch, entertaining myself by doodling molten penises and happy faces in it.
Best day at work EVAR.
Re:Nuke it from orbit by perpenso · 2011-02-17 09:09 · Score: 1

Or in a microwave. That seems to destroy the gates on the chip. 10 seconds on High should be enough. Just be sure to only place the PCB and not the entire drive as they can contain lots metal.
And why can't an attacker just attach a good PCB from a different drive of the same make/model? Assuming of course that the attacker is targeting you specifically and is not just a dumpster diver / recycler who sees a drive and wonders if it works and what is on it. Just removing and breaking the PCB is fine for the later. Although it wouldn't hurt to repeatedly drop drives from 6ft onto concrete until they land flat and rattling noises begin to come from inside the drive.
Re:Nuke it from orbit by lgw · 2011-02-17 09:11 · Score: 4, Insightful

TrueCrypt volume inside a TrueCrypt volume
You, dawg, I heard you liked TrueCrypt.
The headline should just read "Confidential data not safe on unencrypted disk". Modern hard drives also arean't as easy to 100% delete as one might think - once a sector gets "spared out" there's no easy way to delete it, and there will still be readible data there. That just happens a lot less frequently than SSD load/wear balancing.
Of course, any media can be adequetly destroyed by shredding - if you really care, this isn't a problem to solve with software.

--
Socialism: a lie told by totalitarians and believed by fools.
Re:Nuke it from orbit by vadim_t · 2011-02-17 09:13 · Score: 1

Because it's a SSD, and on those the data is on flash chips soldered to it.
Re:Nuke it from orbit by Seraphim1982 · 2011-02-17 09:15 · Score: 1

Because on an SSD the data goes with the PCB.
Re:Nuke it from orbit by h4rr4r · 2011-02-17 09:19 · Score: 1

Removing solder from flash chips is not a huge challenge.
Re:Nuke it from orbit by Calydor · 2011-02-17 09:22 · Score: 4, Funny

This is when you tell us he re-assembled it, loaded up some Linux Drive Recovery program and pulled all the data to safety, right?

--
-=This sig has nothing to do with my comment. Move along now=-
Re:Nuke it from orbit by vadim_t · 2011-02-17 09:24 · Score: 3, Informative

And what good is that?
Again, this is a SSD, not a hard disk. The PCB contains both the interface and the data storage parts. If you microwave that, you've destroyed everything that was important. It's no use to unsolder anything, the flash chips themselves are destroyed by microwaving.
The part you would skip on microwaving is the metal casing, which contains no data.
Re:Nuke it from orbit by trentblase · 2011-02-17 09:33 · Score: 2

The headline should just read "Confidential data not safe on unencrypted disk"
The headline should just read "Confidential data not safe"
Re:Nuke it from orbit by d6 · 2011-02-17 09:34 · Score: 2

should have called the vendor and done an RMA on it afterwards.

"It just quit working, I don't understand..."
Re:Nuke it from orbit by Gaygirlie · 2011-02-17 09:42 · Score: 1

As any self-respecting geek should know the real question isn't whether or not to shoot it with a really big weapon; instead, the real question is whether to shoot at it with one, or use it as ammunition on one.
Re:Nuke it from orbit by perpenso · 2011-02-17 10:03 · Score: 1

Because on an SSD the data goes with the PCB.
Yeah. I had some mental image of the memory being inside a sealed enclosure with an external PCB only handling the interface (SATA, etc). Now that I take a closer look I see that the enclosure is not sealed and that the interface and memory are all on the same PCB. Make more sense for cooling and of course cost reduction.
Re:Nuke it from orbit by GrumblyStuff · 2011-02-17 12:41 · Score: 1

This reminds me of a old post around here. I forget what the topic was or what the site linked to was but I certainly remember the mishmash of computer cables spliced onto standard outlet plugs. I wonder how effective it would be to just start jamming 120v AC power into SSDs.
Re:Nuke it from orbit by Jarik+C-Bol · 2011-02-17 14:43 · Score: 1

Honestly, that nearly amounts to the answer. If you have personal information on ANY drive and are paranoid enough to worry about it to the degree that you think someone with the tech to do recover it after that degree of erasing, Then render the thing down into its base elements with the best cleaning agent ever; Fire. Lots and Lots of Fire.

--
I've decided to Diversify my Holdings. I've divided my cash between my left and right pockets, instead of all in one.
Re:Nuke it from orbit by davester666 · 2011-02-17 15:07 · Score: 1

Yup. Once the confident info becomes a thought in your brain, all bets are off as to who may wind up with access to it.

--
Sleep your way to a whiter smile...date a dentist!
Re:Nuke it from orbit by Meski · 2011-02-17 16:33 · Score: 1

Wuss. *Real* engineers would use 240VAC. Or run it across 2 phases for 415VAC :^)
Re:Nuke it from orbit by Meski · 2011-02-17 16:35 · Score: 1

Fire it into the sun.
Re:Nuke it from orbit by hairyfeet · 2011-02-17 20:42 · Score: 1

Yes lets create even more waste by shredding everything, that's the ticket! As I tell my customers who are nice enough not to buy into the "destroy everything!" bullshit and let me have old machines to refurb for the poor "after a standard DoD 3 pass the ONLY way you are gonna get any data off that thing is to pay some data recovery firm (like in TFA) thousands of dollars to get maybe, just maybe, 10-15Mb worth of data back off a whole disk. Who is gonna waste that much money just for a shot at a tiny amount of YOUR data?"
Is that 10Mb of data which may or may not be recoverable and which will cost out the ass for anyone to even attempt REALLY worth creating ever more waste instead of recycling? In this economy there is a whole hell of a lot of poor folks out there that have nothing, and a working computer can help them post resumes, pick up new skills by taking some of the many free e-classes, and generally just make their life a little better.
Is what you are doing REALLY so top secret that someone will spend the kind of time and money required to get it? If not you are just destroying working hardware for no good reason. There are plenty of guys like me that are happy to refurb machines to give to the poor, but we can't afford to be replacing hardware you trash, so any with drives missing end up stripped or in the dump. Is that REALLY better than just doing a DoD 3 and letting someone have it that needs it?

--
ACs don't waste your time replying, your posts are never seen by me.
Re:Nuke it from orbit by arivanov · 2011-02-17 22:57 · Score: 1

Not from a Flash drive and not from any of the next gen drives which will use overlapping recording combined with Flash cache.
A flash drive has significant excess capacity to ensure acceptable sustained writing speed. When you do a DOD erase it will allocate from that pool and your data will end up unmolested in the spare pool. Unless your OS supports trim or special manufacturer supplied tools that data may linger in the pool for a considerable amount of time. The fact that you are doing a wipe can be at your disadvantage here because the controller may decide to not use any of the pool until you stop abusing it with a sustained write.
So if you want to deal with it at DIY level you do not just need a DOD wipe. You need DOD wipe with scheduled pauses to give the controller time to breathe and wipe the blocks that have gone into the spare pool.

--
Baker's Law: Misery no longer loves company. Nowadays it insists on it
http://www.sigsegv.cx/
Re:Nuke it from orbit by WorBlux · 2011-02-18 05:27 · Score: 1

10-15 MB of credit card numbers and names, is more than enough to do serious damage to your customers. A bit of feel-good isn't worth the risk if the data is critical or sensitive.
Re:Nuke it from orbit by lgw · 2011-02-18 06:01 · Score: 1

The aluminum from a shredded HDD recycles just as well as from an intact one - it's all the same in the furnace. Not that any non-hippie cares about such a bullshit issue to begin with.

--
Socialism: a lie told by totalitarians and believed by fools.
Re:Nuke it from orbit by hairyfeet · 2011-02-18 11:29 · Score: 1

Except you seem to be missing one little important fact: We are giving the machines to poor folks not guys from black hat. And before anybody says "It'll end up on eBay!" no it won't, I've been refurbing machines for the poor and have NEVER seen one end up on eBay, because number one poor folks don't do eBay and number two as long as something works poor folks hang onto it.
I have 400Mhz-600MHz PCs still out in the field being used after all these years and what happens is if the parent happened to get something better the machine gets passed down to the child. Hell one 1.1GHz Celeron that has come through here for little repairs and upgrades a half a dozen times has passed through about 7 relatives in a single family so far, going from mother to child to cousin to aunt etc.
So there is NO NEED to go around creating more waste and destruction when it simply isn't required. in this economy there are literally millions living hand to mouth that could use any help they can get. I'm proud to say most businesses in my area haven't subscribed to the "ZOMG Destroy everything!" attitude and thanks to them I'm looking at some nice socket 478 P4s I'll be refurbing this weekend to give away to those that don't have computers. I have helped churches, a battered women's shelter, and more individuals than I can count...maybe a little story will help change folk's minds about destruction...
I got a call from an old teacher friend of mine a couple of years back asking if I would help him out as a personal favor and do a little free work for this girl taking night classes. The poor little thing was working days and taking early evening classes while her mom watched the kids in order to make a better life for her family. So I get there and the poor thing is in tears because she can't get the software she needs for class to work and Shaun is telling her "It'll be okay, Kevin is a wiz, he'll fix it".
So I get there and the poor little thing is trying to use a 486SX to do her schoolwork on. She looks up at me crying and says "Can you fix it, or at least make it so my son can use it while I try to get another one somewhere?" and I just smile and pat her on the shoulder and say "No problem ma'am, in fact I'll have you fixed up in under 20 minutes" to which Shaun looks like I'm a miracle worker or full of shit, which he can't decide.I tell her to just follow me out to my truck and to back her car up next to it. Lucky for her I'd just got done finishing up a job modernizing a company that was nice enough to let me DoD 3 the old P4s and gave me them along with the monitors/keyboards/mice so I could give them to the poor.
So I sat her box by my truck and opened up the back doors and started to load a couple of nice office P4s into her little hatchback "Oh no" she says "I can't afford to buy any computers right now" and when I told her they were 100% free and that she and her son were getting a couple of nice office machines I thought she'd never stop crying. I took the more powerful of the two into the class and loaded her software along with handing her a little autorun CD I make that has all the software like Firefox and free AV that you need for the basics and sent her on her way. I heard later she graduated with ease and got a really good office job, thanks to having those PCs which she is still using both her and her son from what I was told.
So please don't fall for the hype, poor folks ain't black hats, hell most of them are lucky if they can use a search engine. And with so much eWaste being created yearly and so many poor folks just barely surviving that machine could do some real good out there in the hands of someone needy.

--
ACs don't waste your time replying, your posts are never seen by me.
Re:Nuke it from orbit by WorBlux · 2011-02-18 13:55 · Score: 1

"ZOMG Destroy everything!".
Not everything, but anything where 10-15 MB of data could bankrupt you. If you've set up the system properly, this should only be a small fraction of the machines you have.
Re:Nuke it from orbit by anton_kg · 2011-02-19 23:24 · Score: 1

Absolutely agree. The research has been done using bad initial conditions. They used both Windows and SSD which doesn't support TRIM command. Then they used tools which were not design neither to deal with SSD secure delete ATA command no TRIM command. They have got false results. I guess students don't study hard these days.

My secure erase method still works! by MetalliQaZ · 2011-02-17 08:16 · Score: 3, Funny

1 electric drill, 1 work bench, and some bored interns.

--
"Here Lies Philip J. Fry, named for his uncle, to carry on his spirit"

Re:My secure erase method still works! by TooMuchToDo · 2011-02-17 08:19 · Score: 1

Actually, for an SSD, I'd suggest incineration as a secure wipe method.
Re:My secure erase method still works! by loshwomp · 2011-02-17 08:42 · Score: 3, Funny

And here I thought you were going to bore holes in the SSDs. Boring holes in the interns is just cruel.
Re:My secure erase method still works! by Barny · 2011-02-17 10:56 · Score: 1

Its military standard for erasure so why not?
Would take less than a handful of thermite (or thermate if you have that handy) to securly erase one of these little things.

--
...
/me sighs
Re:My secure erase method still works! by danomac · 2011-02-17 12:16 · Score: 1

Mine works well. Couple of Torx screwdrivers, a 3" belt sander, 50 grit sandpaper, and a vise. It's pretty watching the sparks fly...
Re:My secure erase method still works! by Darinbob · 2011-02-17 13:06 · Score: 1

But they're getting 2 units for college, they should be grateful!
Re:My secure erase method still works! by Thing+1 · 2011-02-17 14:26 · Score: 1

Uh, no, the interns were already bored. :) That's why they just sit there, looking stupid.

--
I feel fantastic, and I'm still alive.
Re:My secure erase method still works! by __aaxtnf2500 · 2011-02-17 15:47 · Score: 1

All classified media generated by the military is destined for burn. Regardless of what you read on some amateur "secure-wipe" program you downloaded as share-ware, DOD does not ever downgrade the classification of media. The people who write those standards keep up on media technology, and as soon as it became impossible to securely wipe a COTS hard disk, they removed the ability to downgrade them. Incineration is what the pro's do. Rending to powder will work fine for those of us without access to high-temperature furnaces, plasma cutters, or welding torches.

Blend it... by Goffee71 · 2011-02-17 08:17 · Score: 2, Funny

... try reading anything from the ensuing dust.

--
If he's the Walrus then can I be a penguin please?

Re:Blend it... by gknoy · 2011-02-17 08:41 · Score: 1

I am absolutely certain that it will blend. If an iphone, ipad, skiis, and a high grade camera suite will blend, I'd be thoroughly surprised if an SSD couldn't. I'm still looking for something they blended with a metal case, though.
http://www.willitblend.com/videos.aspx?type=unsafe
Re:Blend it... by noidentity · 2011-02-17 09:05 · Score: 1

Bah, Chuck Norris would still be able to recover the data just by smelling the dust.
Re:Blend it... by Golddess · 2011-02-18 03:33 · Score: 1

Confidential data dust. Don't breathe this.

--
"I'm not sure I like the fugnutish tone you used in your post!" -RogL (608926)-

How about by Anrego · 2011-02-17 08:17 · Score: 4, Insightful

Encrypting it?

Is taking data off really an issue anyway. If it's confidential data, destroy the disk when you need to dispose of it. Not repurposing or re-selling hardware with sensitive information on it sounds like a no-brainer.

Re:How about by initdeep · 2011-02-17 08:19 · Score: 4, Funny

STOP USING LOGIC ON /.
Re:How about by timeOday · 2011-02-17 08:35 · Score: 1

I don't know why all vendors haven't adopted hardware full disk encryption. This has become an absolute must in my opinion. And compared to software-based encryption, it works so well, and seamlessly - the bios asks for the passphrase at boot time, and after that it's transparent to the OS and doesn't degrade performance either. I would certainly appreciate some security researchers throwing their efforts into validating or debunking these.
Re:How about by Private+Baldrick · 2011-02-17 08:40 · Score: 1

Encryption is a solution that would work. However one of the main reasons for people using SSDs is the performance boost. (Software) Encryption would give a (slight) overhead which might cancel out the benefits of the drive.
Also you need to encrypt the drive FROM THE START. Once data is put down unencrypted in the drive it is potentially retrievable even if you've then encrypted the volume on top.
(Hate to use wikipedia as a resource but... http://en.wikipedia.org/wiki/Solid-state_drive#Comparison_of_SSD_with_hard_disk_drives )
So your encryption choices are either software encryption right from the first build or an SSD with hardware based encryption built in (which is available but not sure how common it is).

--
I have a cunning plan...
Re:How about by mysidia · 2011-02-17 08:41 · Score: 1

Instead of using a SSD use an array of SSDs; with array pairs randomly chosen from a massive pile.
When writing a block, XOR it by a random number of equal size.
Write the random bits to one SSD, write the XOR'ed result to the other SSD.
Then the data cannot be derived from either SSD alone, and neither alone gives you any better chance of getting the data than if you just had a bunch of random bits.
I call this RAID -1 (RAID negative 1), or the opposite of redundant mirroring. That is... without both disks, you have nothing.
Now then... you can call one disk the 'data disk' and the other disk the 'key disk'. The key disk can be written to one time, during initial provisioning of the array, and never needs to be written to again. Therefore instead of being an actual "SSD" the "Key SSD" can be some type of inexpensive read-optimized write-once memory.
It could even be a little chip that gets plugged into the "data SSD" and acts like a read-only secondary hard disk.
When you need to replace your primary SSD because it has run out of write cycles, that's no problem... just DD it to a shiny new SSD, and insert. Put the same "key SSD" in place. All you need to do is ensure the "key SSD" is at least as large in size and has at least as many random bits as the data SSD has capacity.
The old data SSD is worthless and completely unreadable without the "key disk", and there's no reason to destroy it, as long as the "key SSD" remains secure
As for the "key SSD"... there is no reason to ever get rid of it, since it can be re-used without issue, it will never run out of write cycles, because it's never written to under normal operation. Since it can basically be a big hardwired ROM chip, using inexpensive burned fuse links, write-once PROM, it probably costs $10 or less.
If you do choose to get rid of the key SSD... you can make the chance of someone ever putting together the data ssd and the matching key SSD less than lottery odds.
If it's a cheap key device, just melt down the key device and send it for recycling, with minimal ecological damage compared to scrapping large volumes of SSDs.
Re:How about by Guspaz · 2011-02-17 08:41 · Score: 2

SandForce SSD controllers encrypt all data as it hits the SSD. That does nothing to protect against plugging the drive into a computer and using it (a secure delete would handle that), but it *does* protect against people accessing the NAND chips directly. That and the fact that SandForce drives use compression/deduplication/other tricks and properly support secure erase would make it exceedingly difficult to recover data.
Re:How about by camcorder · 2011-02-17 09:00 · Score: 2

One of the SSDs out of 8 they tested has built in encryption. And according to article although it's faster to sanitize data (ie. sanitizing the encryption key), since leftover encrypted data might leave cryptanalysis options, it might be insecure. (though theoretically it is.)
Re:How about by camcorder · 2011-02-17 09:04 · Score: 1

What if you want to delete some portion of data, but still want to use the drive? If it's only one file that you need to get rid of, it doesn't sound like a brainer to destroy all the media, does it? Read the article, it's mostly about this kind of usage.
Re:How about by noidentity · 2011-02-17 09:08 · Score: 2

A problem with full-disk encryption is that it's hard to verify that it's really encrypted on the disk. You have to trust that the manufacturer didn't cut corners and just fake encryption, or botch implementation.
Re:How about by Lord+Ender · 2011-02-17 09:12 · Score: 1

The performance hit is too much. On my netbook, encrypting my SSD brings everything to a crawl. Even on a faster computer, full disk encryption means killing IO performance because everything needs to be routed through the CPU, rather than being able to block-copy data from disk to memory (DMA).
On top of that, to have even 128-bit disk encryption, you need a minimum password strength of 20 random characters. Nobody does this. Nobody. If you picked your password without the use of a random number generator, it could be attacked in better-than-brute-force time.
Now... if somebody could put a bluetooth chip under my skin, and link it directly with my hard drive (so that encryption doesn't tax my CPU), that would make it safe and fast... but ... wait... off to the patent office...

--
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
Re:How about by hawguy · 2011-02-17 09:17 · Score: 1

Instead of using a SSD use an array of SSDs; with array pairs randomly chosen from a massive pile.
When writing a block, XOR it by a random number of equal size.
Write the random bits to one SSD, write the XOR'ed result to the other SSD.
Then the data cannot be derived from either SSD alone, and neither alone gives you any better chance of getting the data than if you just had a bunch of random bits.
I call this RAID -1 (RAID negative 1), or the opposite of redundant mirroring.
That is... without both disks, you have nothing.
You had better choose your random number generator wisely, or your data may not be as secure as you think.
Why not just use a proven encryption method to encrypt everything, then even if someone acquires all of your drives, the data is still secure. (assuming proper key management, of course).
Re:How about by mysidia · 2011-02-17 09:31 · Score: 2

You had better choose your random number generator wisely, or your data may not be as secure as you think.
You don't have to use a "random number generator". You can capture truly random values, since you only have to do it once.
For example, you can hook up a USB geiger counter, place it near a decaying radioactive sample, and collect values measuring the nanosecond timings between photons triggering the counter.
Why not just use a proven encryption method to encrypt everything, then even if someone acquires all of your drives,
There is no such thing as proven encryption. Any encryption algorithm is subject to possibly being broken. And having vulnerabilities. It is possible someone has already broken AES and can decrypt any ciphertext, but they have just not revealed that fact to the public yet.
Entropy methods of masking data involving truly random values cannot be "cracked" by brute forced by finding a hole in a cipher, because there is no cipher.
the data is still secure. (assuming proper key management, of course).
The data is more secure than if it had been plaintext. The data is less secure than if the drive had been destroyed
With a 'key drive' or 'one time pad' that is not disposed with the drive, the data is just as secure as if the drive had been destroyed, assuming a valid implementation of the OTP method.
Re:How about by Anonymous Coward · 2011-02-17 09:36 · Score: 2, Insightful

Congrats... I think you've just built a hard drive sized One Type Pad. Unless I'm missing something, this is horribly vulnerable if an attacker can "borrow" your disk at multiple snaps in time (as you are essentially reusing the one time pad everytime you delete/modify files, so the attacker can now calculate oldfile xor newfile ... if I know at snapshot one I hadn't sent you somebigimage.jpg and at snapshot two, I can basically xor my two snapshots and then xor somebigimage.jpg over all the non-zero fragments and see what file you deleted that freed up space for somebigimage (oversimplified but still and issue)
Re:How about by Private+Baldrick · 2011-02-17 09:46 · Score: 1

Agree completely (and I should have RTFA better rather than skimming it); any volume of encrypted data is vulnerable to cryptanalysis. You only defense is time and strength of the encryption.
If the data is really sensitive (should never see the light of day) then destruction IS the only defense. If the data is sensitive but in a few years will be meaningless (commercial details etc) then most modern crypto should suffice to keep it secure for the next few years if configured correctly.
I think the main fear is with SSDs entering mainstream then data that shouldn't be publicly available is being stored on them (personal details from banks and government) that might be lost/stolen with the laptop.

--
I have a cunning plan...
Re:How about by fermat1313 · 2011-02-17 09:56 · Score: 1

I don't know why all vendors haven't adopted hardware full disk encryption. This has become an absolute must in my opinion. And compared to software-based encryption, it works so well, and seamlessly - the bios asks for the passphrase at boot time, and after that it's transparent to the OS and doesn't degrade performance either.
It has some uses, but for corporate use (where the vast majority of hard drive encryption is done, this is an inadequate solution. There is no provision for backing up or escrowing encryption keys to a central corporate store. Sure the IBM drive supports a user and master password, but this simply isn't workable when you have hundreds or thousand of computers out there with encryption passwords with no central management. Does this support password policies for length, complexity, and aging? I can't say for sure, but probably not.
This is why we currently use McAfee (formerly SafeBoot) at work, and when we transition to Windows 7 this year, we'll use Microsoft Bitlocker. Integration with central directories for password and key management is an absolute requirement for a company of any size.
Re:How about by rgmoore · 2011-02-17 10:00 · Score: 1

You don't have to use a "random number generator". You can capture truly random values, since you only have to do it once.
Great, except that it's going to take ages to collect enough random data for a whole disk. I've tried pre-filling my disks with pseudo random data before doing full disk encryption- the recommended approach, since it makes it hard to figure out where the files are- and it's very slow. On my Core i7, /dev/urandom generates pseudo-random numbers at maybe 500 GB/day. A true RNG will be much, much slower. It might possibly work if you're only trying to encrypt a thumb drive, but it would be hopeless for a full size SSD.

--
There's no point in questioning authority if you aren't going to listen to the answers.
Re:How about by DavidTC · 2011-02-17 10:13 · Score: 1

A part of the system being slower does not automatically make the system slower. It depends on how fast the rest of the system is.
And all O(n) means is 'the process takes a linear amount of time'. I'm not sure what point you're trying to make. Merely copying the data from one place to another is O(n), and the drive does that at least twice.

--
If corporations are people, aren't stockholders guilty of slavery?
Re:How about by DavidTC · 2011-02-17 10:19 · Score: 1

Um, if the attacked 'borrows' one disk, he can't figure anything out. If he borrows it a dozen time, sure...but how?
More importantly, what the hell are you talking about? Instead of borrowing the same thing a dozen times, why doesn't he just borrow both?
Reusing a one time pad is bad if, and only if, an attacker could see more than one output for it.
It's perfectly fine to keep reusing a one-time pad if the previously encrypted message did not, and could not have been, and could not be in the future, intercepted.
It only counts as 'reuse' if it's intercepted twice, not if it's used twice, especially when, like this, the old message overwrite the new.

--
If corporations are people, aren't stockholders guilty of slavery?
Re:How about by zero0ne · 2011-02-17 10:22 · Score: 1

Truecrypt supports hardware acceleration.
(may not be helpful to you with a netbook, but others should know)
Re:How about by DavidTC · 2011-02-17 10:29 · Score: 1

Someone else mentioned this, but it's worth pointing out that in actual real life, large key AES encryption is not breakable. Period. It's not really up for debate. (And if it was up for debate, we've got a lot bigger problems than secure computers drives. Like actual radio transmissions.)
So you do not need an entire drive, you just need a tiny 32meg flash drive with the key on it. Cheap as dirt to have made.
I've actually suggested exactly this before for the NSA when we were talking about them destroying drives. They shouldn't do that. Instead, every computer should be set to boot up and take the key off a locked-in flash drive. Some sort of internal or locked USB enclosure. Or an SD card.
Then, when reusing or even selling the computer, you don't even need to wipe the computer. You just destroy the flash drive.

--
If corporations are people, aren't stockholders guilty of slavery?
Re:How about by Lord+Ender · 2011-02-17 10:46 · Score: 1

This is nice if your hardware supports it, but it still doesn't help with I/O.

--
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
Re:How about by timeOday · 2011-02-17 11:07 · Score: 1

The drive has an onboard cryptographic chip. There is practically no performance impact on the rest of the system. I suppose the drive must consume a little more power than it would without the chip, but I don't think it's much.
Re:How about by hawguy · 2011-02-17 11:10 · Score: 1

You had better choose your random number generator wisely, or your data may not be as secure as you think.
You don't have to use a "random number generator". You can capture truly random values, since you only have to do it once.
I think when you say "once", you mean, perhaps "1 trillion times" if you're filling up a 1 TB hard drive.

For example, you can hook up a USB geiger counter, place it near a decaying radioactive sample, and collect values measuring the nanosecond timings between photons triggering the counter.
How will you certify this home-built apparatus to ensure that the numbers it generators are truly random and there are no biases in your nanosecond timer due to EMF interference or temperature fluctuations that could be exploited to attack the randomness of the data? Building a perfect, unbiased, random number generator (even if based on hardware) is not trivial.

Why not just use a proven encryption method to encrypt everything, then even if someone acquires all of your drives,
There is no such thing as proven encryption. Any encryption algorithm is subject to possibly being broken. And having vulnerabilities. It is possible someone has already broken AES and can decrypt any ciphertext, but they have just not revealed that fact to the public yet.
Sorry, I should have said "industry standard", not proven. Most organizations that need to encrypt data, want to have something that will survive a security audit, and a home-brew encryption method likely will not.

Entropy methods of masking data involving truly random values cannot be "cracked" by brute forced by finding a hole in a cipher, because there is no cipher.
Right, but the hard part is getting the truely random data - and refreshing it periodically so you can do "encryption key changes" which you will likely also have to do to survive a security audit.

the data is still secure. (assuming proper key management, of course).
The data is more secure than if it had been plaintext.
The data is less secure than if the drive had been destroyed
With a 'key drive' or 'one time pad' that is not disposed with the drive,
the data is just as secure as if the drive had been destroyed, assuming a valid implementation of the OTP method.
I agree with this -- as long as you have a source of truly random data, your OTP method is unbreakable. The hard part is in getting the random data.
Re:How about by mysidia · 2011-02-17 11:18 · Score: 1

On my Core i7, /dev/urandom generates pseudo-random numbers at maybe 500 GB/day. A true RNG will be much, much slower.
Your i7's ability to compute random numbers has nothing to do with the rate that true random numbers can be collected using suitable hardware. Your 500gb per day corresponds to approximately 5.92 megabytes per second, which is really quite poor, in terms of performance. This is what happens when you try to execute elaborate computational algorithms; that by no mean indicates random numbers cannot be generated quickly.
The largest Enterprise SSDs that exist right now are ~100gb
That means, you need approximately 858993459200 random bits.
To do that in one hour, you need 238609294bps aka 28.4 megabytes per second.
Anyways, at your slow rate of RNG it would only take 4.8 hours.
Not like it will be a problem for the manufacturer of the "key DISKs". Parallelism can be used to produce output at a high rate, and there are high-speed random number generation methods suitable for OTP use.
Re:How about by couchslug · 2011-02-17 11:29 · Score: 1

Strongly agree!
Hardware is trivially cheap, data is expensive, and destroying a hard disk is easy.

--
"This post is an artistic work of fiction and falsehood. Only a fool would take anything posted here as fact."
Re:How about by mysidia · 2011-02-17 11:34 · Score: 1

I think when you say "once", you mean, perhaps "1 trillion times" if you're filling up a 1 TB hard drive.
I mean that you generate all values required for every bit position on the disk exactly once.
And specifically ignoring any risk that is created, because in some cases, someone can reliably recover old versions of a bit position that was XOR'ed with the same key bit position, if a disk sector is ever overwritten with new data (when new random bits on the key disk are not generated for the data being overwritten).
How will you certify this home-built apparatus to ensure that the numbers it generators are truly random and there are no biases in your nanosecond timer due to EMF interference or temperature fluctuations that could be exploited to attack the randomness of the data?
I won't. Who said anything about making this at home?
Right, but the hard part is getting the truely random data - and refreshing it periodically so you can do "encryption key changes" which you will likely also have to do to survive a security audit.
No. I am suggesting you implement your whole drive encryption scheme on top of this. This total scheme is still not secure against someone stealing the computer system; since both the key drive and the data drive are stored in it. I do not suggest this eliminates the need or utility of also encrypting the volume on top of it. But it is certainly a formidable barrier to any attack effort.
Proper place for the drive keying scheme would be a "failsafe" for organizations who are not satisfied that AES256 encryption is secure to just re-sell the drive after formatting.
With the keying scheme and encryption; there is then simply no justification for the ecological impact of destroying hard drives instead of proper recycling.
Re:How about by hawguy · 2011-02-17 11:41 · Score: 1

This sounds like an awful lot of work just to get around buying self-encrypting hard drives:
http://www.hardwarezone.com/tech-news/view/144196
Re:How about by drinkypoo · 2011-02-17 13:53 · Score: 1

It seems like you ought to use MicroSD cards then, so that the absolute minimum amount of material need be destroyed. It's not like they're going to desolder the chips and destroy those. You can get 4GB and a reader for $6.99, slow as crap but speed is not a big issue here. Then you can use it for config backups or something too.

--
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Re:How about by __aaxtnf2500 · 2011-02-17 15:51 · Score: 1

The main fear with anything entering the mainstream is that the mainstream is vulnerable to such sophisticated attacks as email spearphishing, and you can't fix stupid.
Re:How about by __aaxtnf2500 · 2011-02-17 16:01 · Score: 1

Yes, some people do use long passwords and memorize them. Random passwords.
Most use keyfiles. Much more than 128 bits of entropy.
You have benchmarks that demonstrate that full-disk encryption brings everything to a crawl? That's not the experience of others. It depends on the application, surely, but the blanket statement that performance with full disk encryption will bring a desktop system to a crawl is just nonsense.
Re:How about by Pharmboy · 2011-02-17 16:07 · Score: 1

Great, except he is talking about encryption in the hardware, not the software. If you followed the whole thread, you would know that. Doing it in hardware would have better performance, for starters.

--
Tequila: It's not just for breakfast anymore!
Re:How about by mysidia · 2011-02-17 17:26 · Score: 1

Unless I'm missing something, this is horribly vulnerable if an attacker can "borrow" your disk at multiple snaps in time
Not that an attacker "borrowing" the disk multiple times is considered a threat; if you are using only this scheme.
If you want to defend against that, layer whole disk encryption with XEX-AES on top of this scheme.
And use a removable hardware device for the 'key disk' that gets locked up in a vault separate from the vault the hard disk is locked up, except when the uber-sensitive PC is in use.
The importance of physical security should not be underestimated in preventing hard disk 'borrowing'
Re:How about by L4t3r4lu5 · 2011-02-17 21:37 · Score: 1

Point out to me a hardware encrypted USB key which hasn't been cracked yet, either by outright lying to the customer (AES128 the key, XOR the drive), or by idiot coding errors (the FIPS-140-2 certified Stealth MXP USB key vulnerable to rainbow table attack at release due to the word "PwdHashes" just prior to the actual SHA-1 hashes identified what to give to the crack program).

Seriously, please do. I need some for work.

--
Finally had enough. Come see us over at https://soylentnews.org/
Re:How about by AmiMoJo · 2011-02-17 22:11 · Score: 1

Built-in encryption has to be the key to this (no pun intended). You can use something like TrueCrypt but the performance hit on an SSD is pretty high. Even if you have a CPU with AES acceleration it will be noticeable, assuming performance is that much of an issue for you.
Hitachi and Seagate have been making HDDs with AES encryption built in for four or five years now. The only novelty here is that it is easier to read directly from flash chips than it is from a HDD platter, but we knew that already. How long have USB flash drives been around for now?

--
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
Re:How about by Private+Baldrick · 2011-02-17 22:17 · Score: 1

Encryption isn't falling to Moore's Law, it's falling to weaknesses in the encryption algorithm (the bit that mere humans came up with). AFAIK all the recent in-roads to DES, 3DES, WPA with TKIP have been through faults in the encryption.
What I'm saying is that from a weakness being found, within a few years (or even months), you have scripts that can be run on Amazon EC2 and soon the question is how much money do you want to spend on getting the SSD cracked in a reasonable time.

--
I have a cunning plan...
Re:How about by Lord+Ender · 2011-02-18 01:51 · Score: 1

I think you need to reread my comment.

--
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
Re:How about by __aaxtnf2500 · 2011-02-18 02:05 · Score: 1

I think you need to reread my comment.
Re:How about by hitmark · 2011-02-18 02:57 · Score: 1

The "problem" is that these precautions also bites one the ass when Murphy strikes and one have to hire a recovery company.

--
comment first, facts later. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm
Re:How about by Guspaz · 2011-02-18 03:43 · Score: 1

There's a simple solution to that problem: backups.
Considering that it costs thousands of dollars to hire a recovery company for a single drive, and that tape backups (or other types of backup solutions) cost a tiny fraction of that, the solution is obvious.
Re:How about by DavidTC · 2011-02-18 04:02 · Score: 1

I was actually assuming some sort of custom system, and just using 'USB flash' as an explanation. We're talking about the NSA here, at least I was. The NSA can deal with computer chips.
What would be really clever is to have this chip or MicroSD or whatever inside a card reading hardware on the computer that pulls part of the key off the user's ID card, and the microSD or whatever inside just has the rest of the key. And made as one piece of hardware that can be attached to standard computers, in a drive bay or something, and attached to the USB connector on the mother board. (This internal connection keeps the computer from actually having a 'USB port' that someone could copy info to.)
So even if you steal the entire computer you can't get in, because you lack the card, but just in case someone clones the card, and managed to find the (reused) hard drive, they still can't get in, because when you reused the drive you destroyed the microSD other half of the key. And, unlike the passcard, the microSD thing is uncopiable without someone noticing. (Or actually being logged into the computer...but, um, at that point an attacker hardly needs to copy the card.)
When it comes time to reuse the computer, you just pull that thing out out, and take it back to the secure lab, where you can dismantle it to remove the micoSD card at your leisure, and the techs can do whatever they want with the rest of the computer.

--
If corporations are people, aren't stockholders guilty of slavery?
Re:How about by hitmark · 2011-02-18 04:07 · Score: 1

Note the bit about murphy, that could include backups failing on use (iirc, there are corporations and government institutions that have performed backups for decades but never tested those backups for usability once they are needed).

--
comment first, facts later. http://chem.tufts.edu/AnswersInScience/RelativityofWrong.htm
Re:How about by noidentity · 2011-02-18 04:44 · Score: 1

That's why he said it's more efficient to do in hardware: no-op = just as efficient as non-encryption. Since you can't even tell it's not encrypting, everyone wins!
Re:How about by badkarmadayaccount · 2011-02-21 06:01 · Score: 1

Congrats, you just invented the one time pad - there are other space-time trade-off points, I hope you realize?

--
I know tobacco is bad for you, so I smoke weed with crack.

if you really that concerned use a hammer by Dan667 · 2011-02-17 08:17 · Score: 1

no reading anything after you smash it.

Re:if you really that concerned use a hammer by Dogtanian · 2011-02-17 09:41 · Score: 1

Use a hammer? I tried that, but its data capacity was very poor, I'll stick to solid state hard drives thank you very much.

no reading anything after you smash it.
Plus it's very hard to smash a hammer.

--
"Slashdot - News and Chat Sites Deviant". (Click "homepage" link above for details).
Re:if you really that concerned use a hammer by StuartHankins · 2011-02-17 09:59 · Score: 1

You can smash any hammer with a bigger hammer, right?

Treat it like any other secure system by Brett+Buck · 2011-02-17 08:17 · Score: 2

The solution is the same as hard drives in any secure system - use it, and when you are done, destroy it. Say you get 3 years out of an SSD, the cost of replacing it is trivial over the long haul. Nobody serious about security erases conventional platter HDs and hopes that's good enough.

Re:Treat it like any other secure system by TooMuchToDo · 2011-02-17 08:20 · Score: 1

Exactly. When we recycle computer gear (several tons a year), we wipe the drives first but then I go to the recycling/smelting facility and watch them shred the drives (we have an agreement with the vendor). Trust but verify.
Re:Treat it like any other secure system by somersault · 2011-02-17 08:26 · Score: 2

"Trust but verify"? Verification results from the exact opposite of "trust" :p You're right to verify, but saying stuff like that sounds silly..

--
which is totally what she said
Re:Treat it like any other secure system by Onuma · 2011-02-17 08:27 · Score: 1

Well said. Just like destroying COMSEC in the military -- you can have the two privates complete and sign the blocks for destruction, but the supervisor should always be verifying. After all, it is his ass if things turn up missing.

--
What else can happen when an unstoppable force collides with an immovable object?
Re:Treat it like any other secure system by Solandri · 2011-02-17 08:28 · Score: 5, Insightful

From what I've seen, it's not the end-of-life disposal of drives which leads to this type of data leak. It's when a drive dies under warranty and you send it to the manufacturer for a replacement. Since it's non-functional, you can't erase it. Since you need to return it without any signs of abuse for a warranty replacement, you can't destroy it.

The manufacturer usually just fixes it, and sells it as a refurb / sends it out as a replacement drive for others which have failed under warranty. They just do a quick format, or sometimes even don't bother formatting, before sending the fixed drive out. Meaning the new recipient of your old drive has all your data.
Re:Treat it like any other secure system by jittles · 2011-02-17 08:31 · Score: 4, Informative

The lack of security of SSD's is not new! So unoriginal, in fact, that Truecrypt.org doesn't even recommend that you encrypt an SSD drive!
Re:Treat it like any other secure system by causality · 2011-02-17 08:35 · Score: 3, Insightful

"Trust but verify"? Verification results from the exact opposite of "trust" :p You're right to verify, but saying stuff like that sounds silly..
Verification is after-the-fact. Prior to that, the vendor could still do something dishonest like fail to deliver on its promises. You're trusting them not to do that as indicated by your willingness to do business with them in the first place. Verification is an attempt to check against not only dishonesty on their part but also well-intentioned mistakes that wouldn't strictly be issues of trustworthiness.
It's sort of like when I deposit cash at a bank. If I tell them "this is 200 dollars, please put it into my account" they are going to count the money. I don't take that as an accusation that I am trying to deceive them, because it isn't. It's a standard practice because multiple pairs of eyes are more likely to catch both honest mistakes and deliberate deception. That's an example of "trust but verify".
It's not really so silly and it's far less extreme than "I want to be involved in each step of the process so I can watch your every move". That would be distrust.

--
It is a miracle that curiosity survives formal education. - Einstein
Re:Treat it like any other secure system by jp102235 · 2011-02-17 08:37 · Score: 1

Wait...what??? So they(SSD's) lack security, so truecryot reccomends AGAINST encryption? Shouldn't they brcrdccomending the opposite?

--
jp
Re:Treat it like any other secure system by Nadaka · 2011-02-17 08:37 · Score: 1

Truecrypt recommends you encrypt everything... twice. Even your grocery list.
Re:Treat it like any other secure system by Archangel+Michael · 2011-02-17 08:40 · Score: 1

Trust should never be absolute.Trust is an analog scale, not a digital bit.
Trust but verify is prudent behavior. This is why we pull ever Nth item off a production line, to test and verify that it is worthy of the trust we've placed in the process as a whole.

--
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
Re:Treat it like any other secure system by camperdave · 2011-02-17 09:05 · Score: 2

You should pull every $RANDOMth item off a production line because if your production process has a fault cycle that is a multiple of N items long you'd never catch it.

--
When our name is on the back of your car, we're behind you all the way!
Re:Treat it like any other secure system by the+phantom · 2011-02-17 09:10 · Score: 2

If you are that concerned about the security of your data, then you either encrypt all of your data, in which case it probably doesn't matter what happens to the drive after you get rid of it; or you destroy the drive and suck up the cost of a new one (or you are a large customer, and have an agreement with the vendor which allows you to destroy the drive and get a replacement). Security, convenience, or low cost---pick one.

--
Rhapsody in Numbers
Re:Treat it like any other secure system by noidentity · 2011-02-17 09:13 · Score: 1

That would explain that gibberish grocery list I found the other day while shopping. Wish I'd saved it.
Re:Treat it like any other secure system by ArundelCastle · 2011-02-17 09:13 · Score: 1

Verification is after-the-fact. Prior to that, the vendor could still do something dishonest like fail to deliver on its promises. You're trusting them not to do that as indicated by your willingness to do business with them in the first place. Verification is an attempt to check against not only dishonesty on their part but also well-intentioned mistakes that wouldn't strictly be issues of trustworthiness.
And vendors absolutely do fail to deliver. As you say, sometimes they are mistakes rather than malice, but the result is simply a difference in severity of the complaint, not a reduction in damage to the company trying to properly dispose of the records.
In my work, local and national records vendors alike have seen lawsuits. One had a vendor deliberately reselling paper records as recycling instead of incinerating. There is no such thing as paying someone enough money to trust them.
Re:Treat it like any other secure system by necrogram · 2011-02-17 09:21 · Score: 1

Your better OEMs will allow you to subsitute a statement of destruction for a warrentee replacement. If the drive has had an opertunity to have come in contact with anything classified, its wiped, exposed to a bulk magnetic eraser, then a hammer till the platters are in parts. Dont think a magnet will kill and ssd, but my "Air Force Fine Adjustment Tool" will.
Re:Treat it like any other secure system by WuphonsReach · 2011-02-17 09:29 · Score: 2

If you are that concerned about the security of your data, then you either encrypt all of your data, in which case it probably doesn't matter what happens to the drive after you get rid of it; or you destroy the drive and suck up the cost of a new one (or you are a large customer, and have an agreement with the vendor which allows you to destroy the drive and get a replacement). Security, convenience, or low cost---pick one.

Exactly. Large companies generally have agreements to cover this. A lot of them just unscrew the cover plate off the drive that contains the serial number and model number information and only send that back to the manufacturer.

--
Wolde you bothe eate your cake, and have your cake?
Re:Treat it like any other secure system by VorpalRodent · 2011-02-17 09:34 · Score: 1

I'm more interested in the secret, hidden grocery list embedded in unused portions of the grocery list you found. That's where the person kept the REAL list. The rest was something to make you think you found their grocery list if you happened to stop them at checkout and demand the password.

--
Take it to the limit, everybody to the limit, come on, everybody fhqwhgads.
Re:Treat it like any other secure system by MartinSchou · 2011-02-17 10:24 · Score: 1

It's one thing to link to a document, another not to have read it, but to misrepresent the document you link to shows that you are
a) An idiot
b) Trying to spread FUD
TrueCrypt doesn't talk about a lack of security on SSDs. They talk about the problem of wear leveling. True, this is present on pretty much all SSDs (and flash-based devices), but that is not the same thing.
They say:

Due to security reasons, we recommend that TrueCrypt volumes are not created/stored on devices (or in file systems) that utilize a wear-leveling mechanism (and that TrueCrypt is not used to encrypt any portions of such devices or filesystems).
Now, if someone comes along and invents or build an SSD that does not provide wear leveling (not the best of ideas until we start hitting millions if not billions of erase cycles), TrueCrypt would have no qualms about it. They have no qualms about using TrueCrypt to handle USB keys that have no wear--leveling either.
But be honest about their recommendations. And answer me this: Do you know if your USB keys, SD-cards etc. that you store encrypted information on use wear leveling?
If you don't know, then you obviously do not care about the potential security risk this entails (you're more likely to lose a small item than a computer), and thus you wouldn't care about the potential security issue with an SSD either.
Oh, and if you want to be REALLY paranoid about this issue, I suggest you read their section on reallocated sectors. Let me give you the paranoia-inducing bits here:

Some storage devices, such as hard drives, internally reallocate/remap bad sectors. [...] This may have various security implications. For instance, data that is to be encrypted in place may remain unencrypted in the bad sector.
Ye GODS - it seems TrueCrypt recommends you do not encrypt you hard drives either!
Re:Treat it like any other secure system by geekoid · 2011-02-17 11:20 · Score: 1

If you verify, you don't trust them. IT's pretty simple; which is why Reagan was laughed at for it.
You don't trust their competence, their honesty, their knowledge, their process and so on. IN fact, in many cases verification is a good thing because you can't trust anyone to not make a mistake.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:Treat it like any other secure system by geekoid · 2011-02-17 11:22 · Score: 2

For easy of use, be sure to encrypt everything twice with ROT13

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:Treat it like any other secure system by MobyDisk · 2011-02-17 11:28 · Score: 1

That is what it says, but that isn't what they mean. :-( They should edit that page.
They don't mean that TrueCrypt does not secure data on an SSD. What they mean is two things:
1) If you have an existing SSD with existing unencrypted data, that placing TrueCrypt on it will not delete the existing unencrypted data. Instead, you should start with an encrypted partition from scratch. This is because the unencrypted data is still level in the wear-leveling areas.
2) That you can't plausibly deny that you used TrueCrypt because you can never truly wipe the drive.
Their blanket statement:

Due to security reasons, we recommend that TrueCrypt volumes are not created/stored on devices (or in file systems) that utilize a wear-leveling mechanism (and that TrueCrypt is not used to encrypt any portions of such devices or filesystems).

Is silly and overstated.
Please correct me if I misunderstood their meaning.
Re:Treat it like any other secure system by Facegarden · 2011-02-17 13:29 · Score: 1

Trust should never be absolute.Trust is an analog scale, not a digital bit...
That's always been my problem. My trust was accidentally configured as a floating input.

--
Worldwide Military budgets: $2100 billion. Worldwide Space Exploration budgets: $38 billion. Really, world? Really?
Re:Treat it like any other secure system by jittles · 2011-02-17 13:43 · Score: 1

Well the reason they state this is that it is possible an old header to the encrypted volume could remain undeleted and that an attacker could use that old header to decrypt a reencrypted volume. SO if you fired an employee and changed the passfile or password, he might still be able to decrypt the container, for example.
Re:Treat it like any other secure system by bell.colin · 2011-02-17 15:14 · Score: 1

Dell and other server vendors have this options, just have to pay more at the time of purchase and you get to keep the drive that is being replaced.
Alternately you could just file a typical replacement request for a dead drive, wait for it to be replaced, then not send back the original. You will get a bill though for non-compliance in the returns process.
Re:Treat it like any other secure system by __aaxtnf2500 · 2011-02-17 16:04 · Score: 1

This is a Reagan quote, and Reagan himself understood that it was self-contradictory. It is derived from a Russian saying and they are equally aware of the self-contradictory nature of the statement. That is, in fact, the point. http://en.wikipedia.org/wiki/Trust,_but_verify
Re:Treat it like any other secure system by __aaxtnf2500 · 2011-02-17 16:17 · Score: 1

"Due to security reasons"
And you say Truecrypt doesn't talk about a lack of security on SSD's?

Truecrypt suggests you not use devices with wear levelling to prevent cryptanalysis from leveraging correlations between related data blocks encrypted with the same keying information. This directly concerns the capability of Truecrypt to secure your data against cryptanalysis on SSD's.
Re:Treat it like any other secure system by inglorion_on_the_net · 2011-02-17 21:06 · Score: 1

Good points.
I guess if preventing data leaks is that important, it should be acceptable to just eat the cost of storage devices that fail under warranty. Then the procedure becomes really simple: physically destroy every storage device once no longer in use.

--
Please correct me if I got my facts wrong.
Re:Treat it like any other secure system by MartinSchou · 2011-02-17 21:57 · Score: 1

No, I say they aren't saying "DO NOT USE SSD!". They are saying "DO NOT USE WEAR LEVELING HARDWARE".
SSDs are a subset of wear leveling hardware (as I said in my post). But so are pretty much all flash based drives, including USB, memory cards etc.
TrueCrypt are also saying "BE WEAR OF SECTOR REPLACING HARDWARE" - the kind you find in hard drives, yet the parent to my post doesn't seem to have an issue with that for some strange reason.
Re:Treat it like any other secure system by Svartalf · 2011-02-18 02:45 · Score: 1

Nothing like a good 4# engineer's sledge for fixing a security issue, eh?

--
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Re:Treat it like any other secure system by weepinganus · 2011-02-18 06:17 · Score: 1

The parent's link refers to issues stemming from data not being reliably overwritten on a wear-leveling device. Why wouldn't those same concerns apply to any device that transparently remaps bad sectors to a reserved area of the disk? I understand that most writes to an SSD are wear-leveled, and I assume that transparent remapping of sectors on a magnetic HDD are relatively rare, but isn't information security supposed to be based on the worst-case scenario?
Re:Treat it like any other secure system by jittles · 2011-02-18 06:26 · Score: 1

I think a magnetic HDD is the best you can do, given current technology. How could you feasibly perform any better than that? Besides, you need the entire container header to remain intact. This is possible in an SSD, but very unlikely to occur with a single remapped sector.

Encryption by __aardcx5948 · 2011-02-17 08:18 · Score: 2, Insightful

It doesn't matter if you can get hold of ALL of the data, if it's encrypted you're fucked. Nothing to see here, move along.

Re:Encryption by Anonymous Coward · 2011-02-17 08:40 · Score: 1

It doesn't matter if you can get hold of ALL of the data, if it's encrypted you're fucked. Nothing to see here, move along.
That depends on how good the encryption is, whether or not you can recover the decryption key from RAM, or swap, or brute force, guess it, whether or not the machine is powered on and the drive is mounted already when you seize the device, whether or not you can compel the owner of the device to cough up the decryption key (either legally, or illegally), the kind of encryption used and whether or not it has flaws, whether or not the drive started out decrypted and was encrypted later (because then there may still be unencrypted parts on the SSD), and whether or not their backdoors put into the encryption software.
Re:Encryption by amiga3D · 2011-02-17 08:53 · Score: 1

What's secure encryption today may, a few years down the road, be trivial to break. Best to destroy the drive whether it be mechanical or digital. Most of the time a 3 year old drive is worth a fraction of what it cost new.
Re:Encryption by geekoid · 2011-02-17 11:23 · Score: 1

ALL encryption is possible to decrypt, it just take time.
Of course, beating you until you give me the key also works.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:Encryption by Facegarden · 2011-02-17 13:32 · Score: 1

Actually as someone pointed out above, the wear leveling algorithms in SSD's can cause problems for effective encryption.
http://www.truecrypt.org/docs/?s=wear-leveling
So as always, the answer is "Physically completely destroy the drive".
-Taylor

--
Worldwide Military budgets: $2100 billion. Worldwide Space Exploration budgets: $38 billion. Really, world? Really?

Encrypt the data by rcb1974 · 2011-02-17 08:18 · Score: 1

Solution: Don't copy any data to an SSD unless you're copying it into an encrypted volume.

for the truly paranoid by Seggybop · 2011-02-17 08:18 · Score: 2

I thought we'd already agreed that the only way to be really sure that your data is gone is to physically destroy the drive. If you've got data that's really so sensitive that someone's going to spend serious resources to extract it, the actual price of a drive is nothing. Smash it and call it good.

Re:for the truly paranoid by fermion · 2011-02-17 10:23 · Score: 1

Maybe an acid etch instead of smashing. Hard disk is mechanical, so smashing will destroy the components. But SSD are electrical, so have to destroyt the components. Maybe a day in battery acid to remove the packaging, a day in PC Board cleaner to remove any copper, a day in HF wheel cleaner to destroy the silicone and gates.

--
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black

wipes are vendor specific by gad_zuki! · 2011-02-17 08:19 · Score: 4, Informative

I know OCZ has its own wipe utility and I believe intel too. Using wiping software designed for mechanical disks makes absolutely no sense and the results from this study are 100% predictable. Oh your Gutmann wipe pattern for circa1991 MFM drives doesn't wipe SSDs? You don't say! If you needed to securely wipe one, use the proper tool.

That said, it would be nice if there was some standard way of doing this.

Re:wipes are vendor specific by mlts · 2011-02-17 08:26 · Score: 2

What would be nice is to have the ATA erase command standardized, so this can be easily done.
Command gets handed to the drive controller, controller does the erasing the right way, where on a hard drive, it zeroes out sectors, even the ones on the bad sector relocation table, and sectors marked as bad. On a SSD, it zeroes out everything regardless of the status with regards to wear leveling.
Even better would be having the drive controller encrypt all data, storing the key as a value in NVRAM. Then when it gets handed an erase command, it replaces the key stored with one randomly generated.
Even better would be to have the drive controller to have its own free space bitmap. After being zeroed, if a sector is read without being written to, the controller returns just zeroes, regardless of the actual data present. If the sector was written to, the controller marks it as used in the bitmap and then returns the sector's data on subsequent writes. This way, an erase command can be almost immediate (flagging everything in the bitmap as free), and outside of yanking the controller and looking at the platters/cells, there is no way to retrieve the data that was erased. Bonus points if the controller zeroed out data in the background.
Re:wipes are vendor specific by causality · 2011-02-17 08:43 · Score: 3, Funny

Using wiping software designed for mechanical disks makes absolutely no sense and the results from this study are 100% predictable.
If people were never surprised by predictable things the entire news industry would take a nosedive and be reduced to a shadow of its current self. It'd fuck up the economy!

--
It is a miracle that curiosity survives formal education. - Einstein
Re:wipes are vendor specific by mlts · 2011-02-17 08:46 · Score: 1

The problem is that the ATA commands are there, except there are no utilities available or maintained today that can use them. There used to be a tool called HDDErase, but it requires MS-DOS and a floppy drive for use.
Re:wipes are vendor specific by mysidia · 2011-02-17 08:47 · Score: 1

That said, it would be nice if there was some standard way of doing this.
Wouldn't it be even cooler if they made it part of the ATA standard itself, so you could send a single disk command to immediately commit to secure destruction of the entire volume.... such that even if someone powered down in the middle of an erase and powered the drive back up, the circuit boards inside the drive would just continue the committed physical media secure erase rather than respond to any further read commands, or enable any type of inquiries/recovery efforts?
Re:wipes are vendor specific by blueg3 · 2011-02-17 09:09 · Score: 2

That was their question, too, and they address it in the paper.
Re:wipes are vendor specific by PitaBred · 2011-02-17 10:48 · Score: 1

...you do realize that everything is exactly as you say it should be, right? That's pretty much universal with all SSDs that I'm aware of.
How to perform an SSD secure erase to via ATA commands: https://ata.wiki.kernel.org/index.php/ATA_Secure_Erase
I apologize if your post was sarcasm. It's hard to tell on the Internet, and other people may not recognize it as such.

--
My blog. Good stuff (when I remember to update it). Read it.
Re:wipes are vendor specific by owlstead · 2011-02-17 10:55 · Score: 1

And they work perfectly on SSD's. Since flash is erased per block, they are plenty fast. Much *less than a minute* to fully erase my 80 GB G2 from Intel. Used hdparm. They only annoying thing is the that you need the right access (e.g. many USB devices won't support hdparm, or at least these SATA commands).
Re:wipes are vendor specific by Facegarden · 2011-02-17 13:38 · Score: 1

What would be nice is to have the ATA erase command standardized, so this can be easily done.
Command gets handed to the drive controller, controller does the erasing the right way, where on a hard drive, it zeroes out sectors, even the ones on the bad sector relocation table, and sectors marked as bad. On a SSD, it zeroes out everything regardless of the status with regards to wear leveling.
Even better would be having the drive controller encrypt all data, storing the key as a value in NVRAM. Then when it gets handed an erase command, it replaces the key stored with one randomly generated.
Even better would be to have the drive controller to have its own free space bitmap. After being zeroed, if a sector is read without being written to, the controller returns just zeroes, regardless of the actual data present. If the sector was written to, the controller marks it as used in the bitmap and then returns the sector's data on subsequent writes. This way, an erase command can be almost immediate (flagging everything in the bitmap as free), and outside of yanking the controller and looking at the platters/cells, there is no way to retrieve the data that was erased. Bonus points if the controller zeroed out data in the background.
Better still might be to build flash memory chips with a built-in fuse that cannot be reset. Wipe the data (just in case) and then have some command that physically blows the fuse on every actual flash memory chip onboard. Then someone would have to dissolve the chip and somehow repair the fuse just to get to the data, which would have been erased anyway.
That could make one hell of a virus though if expensive SSD's could be destroyed from software alone. Maybe have it be a (clearly labeled!) jumper on the drive that does the fuse blowing.
-Taylor

--
Worldwide Military budgets: $2100 billion. Worldwide Space Exploration budgets: $38 billion. Really, world? Really?
Re:wipes are vendor specific by Facegarden · 2011-02-17 13:41 · Score: 1

Using wiping software designed for mechanical disks makes absolutely no sense and the results from this study are 100% predictable.
If people were never surprised by predictable things the entire news industry would take a nosedive and be reduced to a shadow of its current self. It'd fuck up the economy!
This just in: this morning a FLAMING BALL OF GAS OVER 1 MILLION TIMES THE SIZE OF THE EARTH APPEARED OVER THE HORIZON! IT IS ABSOLUTELY CERTAIN THAT IT WILL ENGULF THE EARTH IN FLAMES AND DESTROY THE ENTIRE PLANET.*
*this is technically true.

--
Worldwide Military budgets: $2100 billion. Worldwide Space Exploration budgets: $38 billion. Really, world? Really?
Re:wipes are vendor specific by mlts · 2011-02-18 03:19 · Score: 1

Sometimes one wants the drive to erase ASAP, and they might not have time to pop a jumper.
Instead, perhaps a compromise system, where the drive can be set in a multiple security modes:
Mode 1: Password + master one. Good enough for most users, and if the user loses their PW, they can call support with a serial number, and with enough proof, get a one time unlock code.
Mode 2: Erase data after x amount of bad password attempts. The drive controller would zap the key it uses, generate another one, and start again.
Mode 3 is what the PP suggests. Have a fuse and a routine on each flash controller that blows, blocking read/write access to the cells, and the controller would just sit there writing alternate 1s and 0s (or whatever pattern) until depowered. Alternatively, have the memory that stores the encryption key be blown out via fuses first. This way, it would take uncapping with a chip fab to even have a chance at getting to the data.

How is that different than spinning disks? by jklovanc · 2011-02-17 08:19 · Score: 1

It is a commonly known fact that the only way to ensure data is never retrieved from a physical disk whether spinning or SSD is to physically destroy the drive. All other methods short of that have flaws and some data can be retrieved.

Re:How is that different than spinning disks? by firesyde424 · 2011-02-17 08:22 · Score: 2

You know, I've never understood this one. If you have written a zero to every sector on the hard drive, including the hidden space, how in the world is it possible to recover any data at all?
Re:How is that different than spinning disks? by Zironic · 2011-02-17 08:27 · Score: 3, Informative

It's because the bits in the harddrive aren't actually binary but rather values that are intepreted as 1 or 0. For instance a value of 0.6 would be interpreted as 1 and 0.4 would be 0.
This means that if you look at the exact value rather then the interpretation you can make a guess at what values it has been before.
Re:How is that different than spinning disks? by Rashkae · 2011-02-17 08:31 · Score: 3, Informative

By scanning the surface of the platter with specialized equipment, it's possible to detect residual magnetization 'around' the area written by the drive head and determine where there used to be a bit. Actually using this technique to recover anything outside of a laboratory experiment (where the drive was only written to and erased with 0's once) is a myth, however. No one does this, not even CTU.
Re:How is that different than spinning disks? by click2005 · 2011-02-17 08:31 · Score: 1

I didn't RTFA but I'm guessing the wear levelling on SSDs messes up the 'every sector' part. Some sectors get wiped multiple
times while others dont get touched. Writing all zeros is also bad as the magnetic fields from previous data can still be read
(not easily but it is possible). Most modern secure wipes do multiple runs of all zeros, all ones and random data many times.

--
I am a free slashdotter. I will not be modded, blogged, DRM'd, patented, podcasted or RFID'd. My life is my own.
Re:How is that different than spinning disks? by gstoddart · 2011-02-17 08:36 · Score: 3, Informative

You know, I've never understood this one. If you have written a zero to every sector on the hard drive, including the hidden space, how in the world is it possible to recover any data at all?
Essentially, residual magnetism and other sciency-bits.
Suffice it to say, simply writing a bunch of zeros doesn't erase all traces of what was on. With old school HDs, you needed to write random data to each location multiple times -- there's a DoD spec for doing it (DoD 5220.22-M).
I believe the article is saying that it doesn't seem to work with SSDs.

--
Lost at C:>. Found at C.
Re:How is that different than spinning disks? by TheCarp · 2011-02-17 08:39 · Score: 1

Even so, has it even been demonstrated in a lab environment on a disk manufactured in the past decade or so? I was under the impression (from other discussions) that the "area around" that which is written has become so small as to render this pretty much impossible.

--
"I opened my eyes, and everything went dark again"
Re:How is that different than spinning disks? by Archangel+Michael · 2011-02-17 08:50 · Score: 1

If you write out 0s to a disk, and the disk EVER read back a 1 because it was 0.6 then the disk has larger problems than what you're suggesting. You couldn't ever rely upon the bits stored. And by "ever" I mean EVER.
The newer drives, if you wrote 0s out, the density of the data on the platter is so high that it is virtually impossible to recover any data. So writing out 0s is and should be acceptable for 99.99% of the drives. If you are that scared of what is on your drive, just put it into a Magnetic Pulse Machine (Degausser) and then grind it up.

--
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
Re:How is that different than spinning disks? by Roskolnikov · 2011-02-17 08:51 · Score: 1

Wear leveling for flash....
my 120GB OCZ disk has 128GB of space, 8 reserved for dead cells and for wear leveling.
so write 120GB of data to the disk (fill it) remove a text file full of passwords, fill the disk.
the result (if all cells have the same number of uses) would/could be that the SSD in the interest of wear leveling will take lower used cells from the reserve
and leave the cells that I just erased unused.
but heres the problem.
1. all secure data should be, well, secure, encrypted or otherwise
2. this makes a lot of assumptions about the state of the drive, its possible but its going to very difficult (if at all feasible) for anyone but professionals from pulling data off.
if your worried about this and choose not to encrypt running a traditional disk wipe with 3 or 7 pass, wear leveling should still scramble the remaining bits so long as you fill the disk.
Chicken Little was right.

--
Unix, an obscure operating system developed by bored researchers in an attempt to get a better game playing experience.
Re:How is that different than spinning disks? by bitslinger_42 · 2011-02-17 08:52 · Score: 2

It is important to note the section on feasibility in that Wikipedia link... Peter Gutmann did the original (public sector) research on recovering overwritten data on MFM hard drives with very low byte densities (by today's standards). Peter revisited the subject and found that a single overwrite pass, even if only zeroing out every bit, was sufficient to defeat the technique on "modern" drives (i.e. drives larger than 15GB and made in the past 5-7 years).
Re:How is that different than spinning disks? by Guspaz · 2011-02-17 08:52 · Score: 1

No, it wouldn't work, but only because SSDs are copy-on-write by nature, and have large amounts of spare space hidden from the OS. However, using an SSD's built-in secure erase functionality, which triggers an erase cycle on every single block of the SSD, would be sufficient; a flash cell with no electrons in the floating gate isn't going to reveal any secrets.
It should be noted that the multiple rewrites thing is only require for "old school" HDDs. Modern magnetic HDs only need a single pass (as referenced by the wikipedia article that you cite).
Re:How is that different than spinning disks? by blueg3 · 2011-02-17 08:53 · Score: 2

This means that if you look at the exact value rather then the interpretation you can make a guess at what values it has been before.
In theory, maybe. In practice, it's simply not possible. The conventional wisdom that you need to overwrite multiple times, or with patterns, or with random noise, or anything other than just a single pass of zeros is nothing but a myth.
Re:How is that different than spinning disks? by BetterSense · 2011-02-17 08:58 · Score: 4, Informative

It IS pretty much impossible, but that's not going to stop people from perpetuating the wive's tale for decades to come.

I actually have seen Magnetic Force Microscopy used as a tech demo to image the bits on a floppy disk. I asked the process owner if it could be used to extract data, and he just rolled his eyes. He said that besides the issues with modern hard drives having bits that are orders of magnitude smaller both in size and in magnetization, it's just impractical to extract any data, which should be obvious since it takes like 10 minutes to image a handful of bits. A handful of bits that could mean anything, and be anywhere on the disk platter, and anywhere in the file system, and which could represent erased or scrambled or encypted data anyway. I think the idea that you could go beyond even that and divine what bits were written "UNDER" the current ones is just fantasy. I have heard rumors that NSA has made purchases of a large quantity of scanning probe microscopes for this purpose, but they could have just been buying some for testing...manufacturing volume for scanning probe microscopes is such that an order of a half-dozen of them would be an overwhelmingly large order.
Re:How is that different than spinning disks? by Chris+Burke · 2011-02-17 09:02 · Score: 1

If you write out 0s to a disk, and the disk EVER read back a 1 because it was 0.6 then the disk has larger problems than what you're suggesting. You couldn't ever rely upon the bits stored. And by "ever" I mean EVER.
Right, which is why that doesn't happen and isn't the technique used.
The point is that just because the disk (correctly) interprets anything over the threshold as a 1, you can still infer additional information about previous writes based on the actual analog value. Remember, the disk is trying to read the most recent, digital value. In forensic analysis, you're not.

--

The enemies of Democracy are
Re:How is that different than spinning disks? by JonySuede · 2011-02-17 09:02 · Score: 1

It is impossible with know tech, but you can't be sure that some unknown tech will not exist at some point. Therefore it is still safer to destroy the disk.

--
Jehovah be praised, Oracle was not selected
Re:How is that different than spinning disks? by gstoddart · 2011-02-17 09:05 · Score: 1

It should be noted that the multiple rewrites thing is only require for "old school" HDDs. Modern magnetic HDs only need a single pass (as referenced by the wikipedia article that you cite).
Well, the DoD still seem to prefer more 'aggressive' techniques, and apparently don't agree with NIST on this (I believe this is what you were referencing):

As of November 2007, the United States Department of Defense considers overwriting acceptable for clearing magnetic media within the same security area/zone, but not as a sanitization method. Only degaussing or physical destruction is acceptable for the latter.
On the other hand, according to the 2006 NIST Special Publication 800-88 (p. 7): "Studies have shown that most of todayâ(TM)s media can be effectively cleared by one overwrite" and "for ATA disk drives manufactured after 2001 (over 15 GB) the terms clearing and purging have converged."
I still prefer serious physical destruction if you really want to be sure there's nothing recoverable.

--
Lost at C:>. Found at C.
Re:How is that different than spinning disks? by Zironic · 2011-02-17 09:12 · Score: 1

It is as far as I know possible to recover after a single pass of 0's, however possible doesn't mean practical. Since you have to do it on a bit by bit basis a good analogy would be that it's possible to count the number of grains of sand in a beach, but you'd have to be insane to actually attempt it.
It would only be practical if you already know exactly which bits on the drive you wanted to extract and the number was fairly low.
Re:How is that different than spinning disks? by Carrot007 · 2011-02-17 09:24 · Score: 1

On a disk by close examination (electron microscope) you could determine all previous states. In theory. However it would not be cost effective.
On a SSD the firmware would move writes around by default for leveling purposes so obviously things would not be overwritten by attempting to write to the same sector. SSD's have no sectors as such it's all because they are pretending to be legacy drives half the time.

--
+----------------- | What is the question!
Re:How is that different than spinning disks? by Anonymous Coward · 2011-02-17 09:29 · Score: 2, Informative

This is a very popular myth, but after hunting for comfirmation a few years ago I came up empty. Even the original author no longer stands behind this assertion. It's widely considered to be debunked: http://www.lawtechguru.com/archives/2009/03/11_multipass_erasure_myth_debunked.html
Re:How is that different than spinning disks? by blueg3 · 2011-02-17 09:30 · Score: 1

Despite the fact that everyone says it's so, that's no longer the case on modern drives. The signal is so weak compared to the noise that you're looking at something like less than a 1% chance per bit of extracting the original bit instead of a random bit -- particularly useless since you don't know which bits it was successful on.
Re:How is that different than spinning disks? by fishbowl · 2011-02-17 09:43 · Score: 1

>there's a DoD spec for doing it (DoD 5220.22-M).
Whereas the DoD *actually* just shreds them. The most discs I've seen were removed from copy machines so that they could decommissioned. (Did YOU know that copy machines had hard drives?)

--
-fb Everything not expressly forbidden is now mandatory.
Re:How is that different than spinning disks? by gorzek · 2011-02-17 09:44 · Score: 1

I think where most (lay) people get confused is that deleting a file doesn't really delete it, it just removes its link from the file system so you can't see it anymore. The data is still there and may get overwritten the next time something is written to disk, but you can't guarantee it and the data could sit around indefinitely. This is pretty straightforward for the /. crowd to grasp (I assume most people here understand at least a little of how file systems work) but it's kind of a mindfuck to tell Joe Blow that "delete" doesn't really mean "delete."

--
Check out my world simulator thingy.
Re:How is that different than spinning disks? by fishbowl · 2011-02-17 09:46 · Score: 1

In a forensics context, you might not need to recover data. It might be sufficient to merely prove that data was written, when, by whom. You might not need to count the grains of sand on a beach, but it might be useful to be able to say that the sand in your suspect's clothing came from a specific beach at a certain time of the year.

--
-fb Everything not expressly forbidden is now mandatory.
Re:How is that different than spinning disks? by fishbowl · 2011-02-17 09:51 · Score: 1

And anyone who is serious about this either has a hard drive shredder or is getting Iron Mountain or somebody to shred them.
We get a fairly small number of drives removed from military vehicles and equipment like copy machines, that we shred. The machine is pretty brutal, it's a 20HP AMS-2000. I'm sure it was breathtakingly expensive; you need a forklift to move it.

--
-fb Everything not expressly forbidden is now mandatory.
Re:How is that different than spinning disks? by camperdave · 2011-02-17 10:28 · Score: 1

Here's a very handwavy description of how it's done. Although the data stored on a disk drive is digital philosophically, physically it is analog: a region of magnetism on the surface of a disk. As such, you've got to deal with things like variations in field density. They say Nature abhors a vacuum, but Nature abhors any abrupt change in a field and will try to smooth it out. Suppose you have pattern 11101011 on your drive. Over time, the magnetic field in the two 0 areas is going to smooth out the lonely 1 stuck in the middle. Maybe after a few weeks, it isn't a full strength 1. Maybe it's only 0.95 strong. Similarly, the 1s are going to pull the zeros up. Maybe they'll wind up at 0.05. So, a 1 isn't always a 1. and similarly, a 0 isn't always a 0. It doesn't matter, the drive has circuitry in it to "round down" anything below, say, 0.4 and round up anything above 0.7.

Now suppose you zero fill your drive. As always when dealing with electromagnetic forces, there will be opposition to whatever you do. Converting a zero to a zero is going to be fairly easy, however converting a one to a zero is going to take more power. At the same time you are writing a zero field to the disk, there is a counter force writing a 1 there. So your region might contain 0.1 instead of a true 0. For most cases, this won't matter. The aforementioned circuitry will round that to zero anyways.

Professional disk recovery people, however, don't use the standard read/write circuitry of the hard drive. They have sensitive, specialized read heads, and they pick up the 0.1 0.1 0.08 0.01 0.07 0.01 0.08 0.1 pattern, amplify it to 1 1 0.8 0.1 0.7 0.1 0.8 1 which can be "rounded" to 11101011.

Having said all that, a single zero fill pass will wipe the data beyond all hope of recovery for people without highly specialized equipment. A few passes of random data and zero fills, and even the pros won't recover anything.

--
When our name is on the back of your car, we're behind you all the way!
Re:How is that different than spinning disks? by PitaBred · 2011-02-17 10:56 · Score: 1

That may have been true in the early 90's. Currently, it's not. From Wikipedia:

On the other hand, according to the 2006 NIST Special Publication 800-88 (p. 7): "Studies have shown that most of today’s media can be effectively cleared by one overwrite" and "for ATA disk drives manufactured after 2001 (over 15 GB) the terms clearing and purging have converged."[1] An analysis by Wright et al. of recovery techniques, including magnetic force microscopy, also concludes that a single wipe is all that is required for modern drives. They point out that the long time required for multiple wipes "has created a situation where many organisations ignore the issue all together – resulting in data leaks and loss. "
Just because there's a DoD spec for doing it doesn't mean it's necessary or that the spec isn't out of date with technology. Or it's just politically useful to keep it around. Scientifically, there's no current way for someone to even theoretically recover any data from a zero'd drive.

--
My blog. Good stuff (when I remember to update it). Read it.
Re:How is that different than spinning disks? by owlstead · 2011-02-17 11:01 · Score: 1

Really old school HD's. There are several articles on the internet that say that for newer drives, even specialised firms cannot retrieve any data that has been overwritten by random numbers. Of course, you've got things like replaced sectors on current drives, so you are still better off using the SATA ERASE commands that by now have become standard.
Re:How is that different than spinning disks? by geekoid · 2011-02-17 11:17 · Score: 1

yes, because while its have gotten smaller, interments haven't gotten better.
sheesh. It is possible to get a list of the data you can get, and it's even possible to infer data with about 80% accuracy.
Of course, you need to be someone who has access to equipment the sensitive, and understand statistics and probability to even begin to infer the data.
Also you need time.
HOWEVER, it s possible, and it gets easier.
I know this because I was fired for misuse of equipment when I did a POC.
Gotta love corporation.
"Hey, I did this think you can make a lot of money with."
" You' fired."
Granted, this was 7 or so years ago.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:How is that different than spinning disks? by Firethorn · 2011-02-17 12:05 · Score: 2

Well, the DoD still seem to prefer more 'aggressive' techniques, and apparently don't agree with NIST on this (I believe this is what you were referencing):
1. We're paranoid
2. We still have old discs laying around. 10GB? Hah! I've seen 40 MB units, still operational, within the last year.
3. We want to be *SURE*, and the human factor is taken into account - we're willing to overkill on modern drives(and modern is relative), in order to make sure the older ones get wiped properly.

--
I don't read AC A human right
Re:How is that different than spinning disks? by Trogre · 2011-02-17 12:59 · Score: 1

This sounds fascinating. Can you elaborate at all on what equipment you were working and what results you obtained?

--
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
Re:How is that different than spinning disks? by Bigjeff5 · 2011-02-17 15:11 · Score: 1

It's a lot less extreme than the parent suggested, it was simply an example. Older drives would have written analog values as low as 0.8, but that's still more than enough to get a reliable "1" from the write.
In a hard disk, when writing a 0 or 1 you are flipping magnetic poles, and you can never get a perfect 1 or 0. In a high tolerance drive, writing a 1 might actually only write a 0.99, but the read/write head in the drive couldn't tell the difference between a 0.9 and a 1, let alone a 0.99, so it's fine because it is definitely not a 0, it's obviously a 1. Same with writing in a 0.
After several writes and re-writes, you get "1's" that are actually 0.95, or 0.98, or 0.92, and likewise with 0's. These levels are predictable based on the previous value for the data bit, and a sensitive enough reader can tell the difference, and calculate the changes, reproducing the data at each change. After a certain point there is too much noise and you can't track the changes any further, because the write head on the hard drive is never going to write anything like a 0.8, it's far more precise than that.
This is why for decades the standard for securely wiping a hard drive has been a pattern of 7-15 writes and re-writes. Usually start with all 0's, then all 1's, then alternating 1's and 0's, then flipped, etc.
The technique relies on the forensic reader being far more precise than the hard drive write head. Because of this the technique has become more difficult as hard drive capacities have increased - the platters are much more dense, and so the write heads have become much more precise. Still, several changes are trackable this way.
NAND flash is different, and this study clearly shows that the above sanitation techniques are not reliable (sometimes they work, sometimes they don't). Techniques need to be specifically designed for NAND flash in order to sanitize SSD drives.

--
Security is mostly a superstition... Avoiding danger is no safer in the long run than outright exposure. - Helen Keller
Re:How is that different than spinning disks? by toddestan · 2011-02-17 15:52 · Score: 2

As someone else who's played around with magnetic force microscopes, recovering data off of a disk would be extremely time consuming. As the parent mentioned, you're talking several minutes to capture an image that's maybe 100 square micrometers (10x10 um). A floppy disk has several million square micrometers of surface area to image per side - you're literally talking centuries to read a disk this way.
The other problem is resolution. I haven't seen a microscope yet that can see the bits on a modern hard drive. If you want to see bits, you're generally imaging a floppy disk, or an old MFM/RLL hard drive. Zip disks also work well.
Of course, it could still be the wrong tool for the job. A $100k magnetic force microscope may take centuries to read a diskette, but a cheap $15 floppy drive can do it in about a minute.
Re:How is that different than spinning disks? by __aaxtnf2500 · 2011-02-17 16:25 · Score: 1

The problem is that the data on your hard disk is organized in magnetic domains that slightly vary in size and location on each write. What appears as a discrete array to the driver and operating system is a fuzzier set of measurements that the disk logic interprets into a well-behaved array.
The way this is exploited to recover deleted data is with a high resolution scanning electron microscope. Nation-state actors possess the technology to reimage a drive platter on microscopic scale and rebuild probable data structures whether they have been erased or overwritten or whatever.
To what extent this is practiced is classified
There are plans you can find through google to build your own microscope for hard disk recovery for less than 10^3 USD.
Re:How is that different than spinning disks? by Svartalf · 2011-02-18 02:57 · Score: 1

You might not be able to extract it with the hardware provided by the disk itself, but the platters CAN be read with other gear and like all other magnetic media, retain ghosts of the stuff you wrote out to it previously unless you use massive alternating magnetic fields to erase the whole media.
For example, if you reused some tape in a reel-to-reel or an audio cassette, you might notice ghosts of previous recordings on your silent parts of the current recording- unless you used a bulk tape eraser (which used AC and a VERY intense for the media magnetic field to wipe the tape...) as the heads had the power to put your current content on there as well as sort-of erase the old stuff. It's little different with hard disks, really, but it's quite a bit harder to "bulk erase" them. Seems there's some similar issues with the SSD's, though I wouldn't have categorized them as being a questionable place to put data as the article submitter presented- it's really much the same problem with spinning magnetic disks and the DoD would demil things by physical destruction if they needed it to stay classified.

--
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas

what, you don't have a firepit? by swschrad · 2011-02-17 08:21 · Score: 1

excellent tool for neutering storage. build up a roaring fire with about 6 inches of coals, and then toss the hard disk into it. retrieve in morning, dump in trash. done.

--
if this is supposed to be a new economy, how come they still want my old fashioned money?

Re:what, you don't have a firepit? by ColdWetDog · 2011-02-17 08:26 · Score: 1

excellent tool for neutering storage. build up a roaring fire with about 6 inches of coals, and then toss the hard disk into it. retrieve in morning, dump in trash. done.
Don't be so sure of that.

And now, data recovery experts announced they were able to salvage scientific data from a charred hard drive.
Said hard drive deorbited on the Columbia.

What NASA sent to Kroll Ontrack was almost unrecognizable as a hard drive. Jon Edwards, a senior clean room engineer at the company said that the circuit board on the drive was burned beyond recognition and that all its components had fallen off. Every piece of plastic on the 400 MB Seagate hard drive had melted, and the chips were burned.

--
Faster! Faster! Faster would be better!
Re:what, you don't have a firepit? by tragedy · 2011-02-17 08:41 · Score: 2

Sure, but the drive casing probably didn't break open. It would have been made of aluminum, most likely, which isn't the best heat sink, but is better than nothing. The heat it was exposed to was probably intense but brief. So, the platters inside the drive were probably only exposed to a small amount of heat for a short period of time. The overnight fire that the grandparent post referred to would be hundreds of times longer and probably hotter too.
Re:what, you don't have a firepit? by JonySuede · 2011-02-17 09:07 · Score: 1

but this was a 400mb radiation hardened disk with magnetic domains a few magnitudes bigger than modern 2TB disk

--
Jehovah be praised, Oracle was not selected

Just don't do secure data by Murdoch5 · 2011-02-17 08:21 · Score: 1

Okay so it's not so secure, for secure data use secure highly encrypted mediums. If you encrypt the data on the SSD does it matter how much is left, if you end up with encrypted data how can anyone use it with no clue on how it was encrypted, for going good crackers and hackers. I'd assume there not pulling off full data, just fragmented data so that's even harder to put together.

thermite will fix that by WhiteDragon · 2011-02-17 08:22 · Score: 4, Insightful

Thermite will fix everything! [s/fix/destroy] :-)

--
Did you mount a military-grade, variable-focus MASER on an unlicensed artificial intelligence?

dd by hwk_br · 2011-02-17 08:23 · Score: 1

Didn't RTFA, but how dding zeros to the device?
dd if=/dev/zero of=/dev/sdb should work on everything...
I remember something about a prize for recovering data from a zeroed HD...

--
\m/

Re:dd by Zironic · 2011-02-17 08:30 · Score: 3, Informative

According to RTFA they can recover almost 100% of the data from a 0'd HD, 90% of the data from a randomed HD and 1-10% from a HD that has run extremely extensive random HD passes (Like Gutmann)
This is due to SDD's working differently then the standard HD's.
Re:dd by Anonymous Coward · 2011-02-17 08:34 · Score: 1

I assure you, a single pass of writing zeros to any drive isn't a secure way of erasing data. Even multiple passes of /dev/urandom will only make it extremely difficult for all but the most determined person(s).
Re:dd by blueg3 · 2011-02-17 09:13 · Score: 2

No, that's only for attempting to perform a secure erase of a single file. The results for trying to secure-erase single files are so bad (and since there is no ATA command to securely erase only particular blocks on a drive) that it is unsafe to write data to an SSD and then hope to reliably remove that data from the drive without zeroing the entire drive.
If you'll RTFA carefully, though, you'll note that for all but one drive they tested, zeroing the entire drive was reliable. One drive had about 1% of the original data remaining after 20 passes. One drive was entirely erased in one pass. The other drives were entirely erased within 2 passes.
So, zeroing an entire SSD works as long as you use more than one pass. Zeroing individual files on an SSD doesn't work.
Re:dd by km_2_go · 2011-02-17 13:41 · Score: 1

If you write all zeros to a drive, I can recover all the data. All the data will be zeros, but still...

truecrypt by SharpFang · 2011-02-17 08:25 · Score: 5, Insightful

encrypt the data before writing. at no point in its existence will it appear anything but white noise to unauthorized parties.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2

Re:truecrypt by pentalive · 2011-02-17 08:54 · Score: 1

The "unauthorized parties" will use a $5.00 wrench to beat you until you tell them the password or as in the case of Great Britain, throw you in jail until you remember it.
Re:truecrypt by SharpFang · 2011-02-17 09:11 · Score: 1

They can do the same thing if the data is not there. They -still- won't get the data just the same.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by Rich0 · 2011-02-17 11:02 · Score: 1

Most full-drive encryption systems used in corporations use TPM to secure the key. No human being knows what it is, although optionally the corporation might retain a copy on a secure server somewhere.
With TPM you can actually encrypt a hard drive such that you can boot it without any passwords normally, but if you remove the drive or boot from alternate media you cannot read the drive. The only way to circumvent it is to attack the TPM chip, which of course is specifically designed to resist any kind of attack. The TPM only gives away the key if requested by the software that generated the key, as certified by a trusted boot chain.
It cannot be perfect, but it is very strong. And, if all you have is the hard drive then the key isn't stored on the drive at all, which does make it as strong as the applied crypto (which still can have weaknesses).
Re:truecrypt by geekoid · 2011-02-17 11:10 · Score: 1

Yes, so now you are beaten, and your assailant is even more angry. How well with that end for you?

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:truecrypt by SharpFang · 2011-02-17 11:37 · Score: 1

Doesn't matter. I'm just a meaningless human. I'm disposable. But THE SECRET is safe.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by trawg · 2011-02-17 12:02 · Score: 1

Interestingly (as noted above) TrueCrypt has a warning on creating encrypted volumes on volumes that do wear-leveling, which I assumes includes most SSDs these days?
http://www.truecrypt.org/docs/?s=wear-leveling
Re:truecrypt by slashqwerty · 2011-02-17 15:18 · Score: 1

encrypt the data before writing. at no point in its existence will it appear anything but white noise to unauthorized parties.
--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2

Your statement is very ironic in light of your signature. Oh, I'm sorry, that was just a random number you posted.
Re:truecrypt by dotancohen · 2011-02-17 19:57 · Score: 1

encrypt the data before writing.
at no point in its existence will it appear anything but white noise to unauthorized parties.
WRONG

--
It is dangerous to be right when the government is wrong.
Re:truecrypt by Asic+Eng · 2011-02-17 21:09 · Score: 1

Depends what your attack scenario is. If you want to protect yourself from the state you need to take different measures then if you just want to protect your trade secrets in case you have your laptop stolen or you lose your USB drive. The latter two scenarios are probably more interesting for the majority of us.
Depending on the jurisdiction you are in, encrypting helps against the state as well. E.g. in Germany you can not be compelled to give evidence against yourself, and that includes giving up encryption keys.
Re:truecrypt by SharpFang · 2011-02-18 00:58 · Score: 1

Did you read the article you linked? Did you read some other posts in this thread?

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by dotancohen · 2011-02-18 03:23 · Score: 1

Yes. TrueCrypt is not effective on SSD drives.

--
It is dangerous to be right when the government is wrong.
Re:truecrypt by SharpFang · 2011-02-18 11:01 · Score: 1

It's called functional illiteracy. You read but you don't understand what you read. Truecrypt might fail on SSD drives in two cases: - the drive contained any sensitive data prior to encryption - you needed to change the volume header due to the old one getting compromised. These are the only two scenarios and both are easily avoided.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by SharpFang · 2011-02-19 08:31 · Score: 1

Note that even then, the classic method of "fill the drive to the brim" - create a file to fill the space 100%, works quite reliably. It -may- not work if new faulty sectors have been found and have been mapped out. Checking for this change is perfectly doable (there are lists of these sectors, clearly readable). Even then the chance this affects you is minimal, because 1) that particular sector must have kept the compromised header and 2) damage to that sector (that caused mapping it out) hasn't destroyed that header data.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by dotancohen · 2011-02-20 09:49 · Score: 1

Either I'm suffering from functional illiteracy, or you are wrong.
Even if the drive contained no sensitive data prior to encryption, the user can never change the password as that will change the volume header. And there need not be worry of compromise to change a password and you imply.
Furthermore, the page states that even if the user does manage to properly encrypt the drive before putting sensitive data on it by using a complicated procedure (which includes disabling the OS swap / paging file and other non-end-user steps), there is _still_ no guarantee of reliable encryption. Even furthermore, even if the goal of encryption is achieved, there is no plausible deniability for encrypted partitions on SSD's.
Still not convinced? Then trust _your_ state secrets to them. And stop with the ad hominim (sp?) attacks.

--
It is dangerous to be right when the government is wrong.
Re:truecrypt by SharpFang · 2011-02-20 18:12 · Score: 1

Functional illiteracy and unwillingness to read replies posted already.
1. Password change still possible and safe as long as no new bad blocks appeared since encryption. Even then the chance the badblock contains still readable header information is minimal. How? Read my other reply. It's slow but it's not a frequently done operation.
2. The same method can be used to remove any leftover data from system installation.
3. The plausible deniability implies you do have OS with truecrypt, and a truecrypt partition/file/storage and this much IS publicly known (as opposed to existence of a hidden partition). And that's exactly as much information as can be recovered from the unencrypted scraps of data. There will be indication of OS and Truecrypt present, but there won't be any indication whether it was used for any hidden partition or not.
Preconditions: the drive is of a good quality and provides a reliable list of its mapped out bad blocks. The moment new bad blocks appear, the password cannot be safely changed any more (although probability the old password remains available is still minimal). Only bad blocks are of concern, wear-protection mechanisms are not.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:truecrypt by dotancohen · 2011-02-20 22:49 · Score: 1

Could you please phrase that as a car analogy?

--
It is dangerous to be right when the government is wrong.

thermite meh by Fujisawa+Sensei · 2011-02-17 08:29 · Score: 1

I prefer a mixture of magnesium dust and gunpowder; but to each their own.

--
If someone is passing you on the right, you are an asshole for driving in the wrong lane.

Re:thermite meh by Svartalf · 2011-02-18 03:02 · Score: 1

Heh... Either concoction will make a hash of the HD/SSD it's used on- and do it rather swiftly.

--
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas

Data recovery by Lorien_the_first_one · 2011-02-17 08:29 · Score: 1

I guess what concerns me the most about SSDs is data recovery. Is that any harder on SSDs than regular disks? Or is data recovery a moot point since there are no moving parts?

--
The diversity and expression of human opinion is essential to human survival.

Re:Data recovery by dgatwood · 2011-02-17 08:40 · Score: 1

Well, it's a wash, based on the last stats I read. (I forget where I read the article.) With SSDs, you have no moving parts, which makes them much, much more reliable in portable devices (laptops, iPods, and so on). However, you have many more solder joints to crack, so you have a much greater chance of a thermally-induced failure than you would with a hard drive.
The real advantage of SSDs as far as data recovery goes is that you don't need a clean room to work on them. The majority of failures in electronics are caused by broken solder joints on the board, which can be repaired by anyone who owns proper reflow soldering tools.
The one place where SSDs are at a disadvantage is that a board failure on a hard drive can be repaired by swapping the boards in many cases, whereas a board failure on a SSD requires actually diagnosing the board, and in the worst case, unsoldering the flash parts and soldering them to a new board. Even still, the whole "anybody with reflow tools" rule makes that not nearly as bad as it sounds.

--
Check out my sci-fi/humor trilogy at PatriotsBooks.
Re:Data recovery by Rob+the+Bold · 2011-02-17 09:05 · Score: 1

I guess what concerns me the most about SSDs is data recovery. Is that any harder on SSDs than regular disks? Or is data recovery a moot point since there are no moving parts?
That's the other side of the data security coin, isn't it? Getting it back after some "unfortunate incident". Wei and Grupp seem to suggest that it's easier, at least how I read it. And it sounds like they're just hacking around the control logic: "we have designed a procedure to bypass the flash translation layer (FTL) on SSDs and directly access the raw NAND flash chips". Whether or not they mean "ICs" when they say "chips", I dunno. Kinda makes a big difference if you've got to saw, pry or etch off the package vs. do a little desoldering.

--
I am not a crackpot.

It is difficult by crow · 2011-02-17 08:32 · Score: 2

You can't do a secure erase from software, because data may still exist in blocks that were remapped by the firmware due to errors or for write leveling. When you write to an SSD, the new data goes in a free block, and the old block is marked free. To do a real secure erase, you have to work with the SSD firmware, and even then, you can't be sure if data may still exist on bad blocks that can't be written to.

So the only way to be sure is to physically destroy it, and flash is reliable enough that it's difficult to be certain that you've truly destroyed it.

So as everyone else is saying, the only good solution is to encrypt everything, and don't store the keys in flash.

Re:It is difficult by h4rr4r · 2011-02-17 09:13 · Score: 1

So the only way to be sure is to physically destroy it, and flash is reliable enough that it's difficult to be certain that you've truly destroyed it.
Thermite, problem solved.

Secure erase option by eagl · 2011-02-17 08:35 · Score: 1

A couple whacks with a hammer still works great. Remove the circuit board from the case, give each chip a little love tap with a ball peen hammer. Problem solved without waiting hours for the thing to "secure erase".

Concerned about losing resale value? Security costs money, period. If you want real security, sometimes you have to take some financial responsibility and accept the loss of resale value in exchange for real security. Price of doing business.

Re:Secure erase option by Guspaz · 2011-02-17 08:56 · Score: 1

SSD secure erases are almost instant. The SSD might not be able to write to every cell simultaneously, but it *can* erase them all at the same time.
Re:Secure erase option by geekoid · 2011-02-17 11:08 · Score: 1

or, you know, design the SSD so you can erases the sector on the fly if you want, instead of allocating a new one and waiting.
Huh, now you get to erase and reuse without destroying something.
I know it may be alien to you, but you don't need to destroy things to solve as many problems as you think.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Re:Secure erase option by Overzeetop · 2011-02-18 01:24 · Score: 1

Wait. You can't change all of the gates to a random state (1 or 0) at one time, but you CAN change them all to a 0 simultaneously? That seems impossible without an alternate "self-destruct" write path to every cell.

--
Is it just my observation, or are there way too many stupid people in the world?
Re:Secure erase option by Guspaz · 2011-02-18 03:42 · Score: 1

You can't change the state of a flash cell without first erasing it, they already have that "self-destruct" write path. When you get an empty SSD, all the blocks are in the erased state, so each write is a simple write. If you want to rewrite anything, you must first erase it before you can write to it. That's why SSD performance slows down over time. A secure erase doesn't do any writing, it just erases all the blocks simultaneously withing having to care about which ones should be preserved.
I'm probably slightly exaggerating by saying that it happens instantly, but the last time I did a secure erase on an Intel SSD (due to getting it into a poor performance state), the process took mere seconds.

Re:Pure crap by eagl · 2011-02-17 08:36 · Score: 1

If you use the proper erase methods (solid state or other) then it doesn't matter. If you need to destroy the data simply put it on a cookie sheet and put it in the over on broil for 30 minutes.

Wifey hates the smell of burning plastic in the oven. Don't ask me how I know this.

Once you put it on... by fahlesr1 · 2011-02-17 08:37 · Score: 1

"Makes you think twice about storing data on SSDs — once you put it on, getting it off isn't so easy."

My 12 gauge begs to differ. Pull!

Technique for recovery by DieNadel · 2011-02-17 08:37 · Score: 1

For once I've read the paper :-)

But I could not find a description of the technique utilized to recover the files.

They say that an "advanced hacker" will be able to recover the files, but I'd like to know how.

--
Utinam logica falsa tuam philosophiam totam suffodiant!

Re:Technique for recovery by compro01 · 2011-02-17 08:51 · Score: 1

Presumably dump the contents of each individual chip.

--
upon the advice of my lawyer, i have no sig at this time
Re:Technique for recovery by blueg3 · 2011-02-17 09:14 · Score: 1

They made an FPGA board that interacts with the flash chips directly, bypassing the translation layer (FTL).

Re:"...getting it off isn't so easy." by jgagnon · 2011-02-17 08:39 · Score: 1

I don't know about any of you and I'd like to keep it that way...

--
Remember to maintain your supply of /facepalm oil to prevent chafing.

I think I'm safe by lxw56 · 2011-02-17 08:41 · Score: 5, Funny

I challenge anyone to find my MicroSD card. I've conducted extensive security audits to verify that no attacker, even one with inside information, can gain electronic or physical access to the disc.

Re:I think I'm safe by BrentH · 2011-02-17 08:45 · Score: 1, Funny

It's up your ass?
Re:I think I'm safe by pipatron · 2011-02-17 08:57 · Score: 1

Apparently yours. It wooooooshed up there.

--
c++; /* this makes c bigger but returns the old value */
Re:I think I'm safe by RapmasterT · 2011-02-17 09:00 · Score: 4, Funny

I challenge anyone to find my MicroSD card. I've conducted extensive security audits to verify that no attacker, even one with inside information, can gain electronic or physical access to the disc.
Translation: "I lost the tiny little bastard and can't fucking find it!"
Re:I think I'm safe by dadelbunts · 2011-02-17 09:10 · Score: 1

I always find it funny how technology has gotten so tiny i can eat 4 gigs with no problem. Dont even have to chew.
Re:I think I'm safe by SharpFang · 2011-02-17 09:22 · Score: 1

8 gigs currently; 16 if you have lots of money to waste.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:I think I'm safe by dwarfsoft · 2011-02-17 10:54 · Score: 1

I have 16. It's 32 that requires the wasting of lots of money.

--
Cheers, Chris
Re:I think I'm safe by SharpFang · 2011-02-17 11:04 · Score: 2

thanks for update.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2

Summary by Orgasmatron · 2011-02-17 08:42 · Score: 5, Informative

Block storage devices have more capacity than they report. Magnetic disks keep a small reserve of unallocated blocks as a hedge against blocks that fail in use. SSDs keep a much larger reserve because they can only erase in increments that are relatively large compared to their block size.

If you overwrite a sector on a magnetic disk, you will almost always destroy all traces of the old data. The exception is when the drive thinks the old sector has failed or is about to fail, in which case you get an entirely new sector, and your old data is still (possibly) on the old sector. Attacks using magnetic force microscopes to read data from track fringes were possible a decade ago, but there is no reason to think it is possible on a modern drive.

If you overwrite a sector on a SSD, the SSD gives you a whole new block from a list of free blocks, and adds the address of the old block to the list of deleted blocks. Blocks are moved from the deleted list to the free list when the SSD has some free time, or when one is really needed. There is currently no mechanism to force the SSD to actually erase a sector.

This is all known, and there are mechanisms built into the specs to provide a secure erase. What their research is showing, however, is that these mechanisms don't always work. A number of them are buggy, and at least one just plain lies, claiming to have done the secure erase, but actually just doing the normal pointer update trick just like any other write.

--
See that "Preview" button?

Re:Encrypt it.... by mosb1000 · 2011-02-17 08:43 · Score: 1

Someone once told me that I should use RSA encryption because it was developed by the NSA. I thought to myself "why would the NSA produce and give away an encryption algorithm they can't break". I concluded that they wouldn't. So yeah, probably not secure.

Because things are really analog not digital ... by perpenso · 2011-02-17 08:46 · Score: 1

You know, I've never understood this one. If you have written a zero to every sector on the hard drive, including the hidden space, how in the world is it possible to recover any data at all?

Because digital is just a convenient abstraction for our analog reality. Here's a gross simplification. A bit is just a magnetic blob on a large plane of magnetic media. When a read/write head returns to a particular spot it does not return to exactly that same position, close but not exact. As the platter spins and it lays down a track of these magnetic blobs it may write the new track a little bit to the side of the old track. This partly motivates wiping software writing data seven or more times, it wants to increase the likelihood of getting the old data.

Try this: Take two hilighters, one yellow and one a darker color. Draw a yellow line. Now draw on top of that line with the other color. See any pure yellow peeking through on the edges? That yellow is like the area where data recovery people will use highly specialized equipment to read "overwritten" data.

Good thing? by LoudMusic · 2011-02-17 08:50 · Score: 1

This sounds like a good thing to me. Better chances of getting data back from failed hardware. Or getting data from a device that a numbskull disgruntled employee thinks they've intentionally ruined.

If you actually WANT to destroy the data, others here have mentioned the proper methods. I like to rely on the .45 at high velocity, but open flames work well too.

--
No sig for you. YOU GET NO SIG!

Re:Overwrite by compro01 · 2011-02-17 08:54 · Score: 1

The problem is that doesn't work due to wear leveling. The virtual area you're overwriting isn't necessarily the same physical area that holds the data you want gone. Even wiping the entire thing doesn't do it, thanks to spare blocks.

--
upon the advice of my lawyer, i have no sig at this time

you mean reading the entrails? by G3ckoG33k · 2011-02-17 08:55 · Score: 3, Funny

You couldn't possibly seriously mean we should start reading the entrails? That is soo medieval.

Umm... by Reber+Is+Reber · 2011-02-17 08:55 · Score: 1

"once you put it on, getting it off isn't so easy." - That's what she said?? ZINGGGGGGGGGGGGGGGGG!

Working as intended by Ralcog · 2011-02-17 08:56 · Score: 1

They werent designed for secure erasure. The article says that varying methods of erasing the drives such as DoD 8 pass still leaves atleast 1% of data that can be recovered.

Presumably this is because of the optimization techniques that SSD's use to achieve high performance and increase lifespans.One of these measures is having 64GB of flash on a 60GB SSD, leaving extra flash to act as for intesive operation and wear leveling. Since the disks werent designed for secure erasure, no method erases the extra space, and what conventional program do is just trigger the controller to sometime overwrite some of the extra flash space.

This isnt endemic to SSD technology, just the way the controllers are implemented. At some point controller will probably support this secure erase of all flash.

TL; DR : new tech doesnt have all the features, can recover atleast 1% of your data until better tech comes out

Re:Working as intended by fishbowl · 2011-02-17 09:28 · Score: 1

I laugh whenever I see references to the DoD specifications for erasure. The way the defense contractor I work for actually handles it is with a shredder. A big ugly 20HP shredder that weighs about two tons and will turn disc drives into confetti that looks like lathe chips as fast as you can feed them in. (The brand is "AMS" if you're interested.)

--
-fb Everything not expressly forbidden is now mandatory.

Just use a more powerful data descrution device. by h4rr4r · 2011-02-17 09:01 · Score: 1

I find 165 gains going about 3000 fps is a very effective data destruction device. It is also a great way to relieve stress.

Not so fast, Batman! by moxsam · 2011-02-17 09:06 · Score: 1

For a system drive you have to at least install the OS before being able to encrypt it with TrueCrypt or its fork DiskCryptor.

That's not a problem if you don't save any personal data to the drive after installing the OS and before a system encryption, but nevertheless this depends on how wide you define personal data. Is the choice of OS, any registry key, choice of software, isn't that personal information, too?

Re:Not so fast, Batman! by SharpFang · 2011-02-17 09:18 · Score: 1

I'm not sure but I think once I have the pre-boot authentication in place, I can install a different OS. And if I'm wrong, the only thing they might learn (with a lot of effort, recovering scraps of data total overwrite failed to remove, as per The Article), what OS I use. No 3rd party software, no registry and the likes. The end. I can encrypt the system volume the first thing after I'm able to run a first program on the system. All the other software, all modifications to the registry and so on, will run on the encrypted disk.

--
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
Re:Not so fast, Batman! by moxsam · 2011-02-17 10:14 · Score: 1

I'm not sure but I think once I have the pre-boot authentication in place, I can install a different OS.
No, sorry. System encryption always needs a device/disk driver in the OS. The decryption code does not run on top of the kernel like a rootkit.

And if I'm wrong, the only thing they might learn (with a lot of effort, recovering scraps of data total overwrite failed to remove, as per The Article), what OS I use. No 3rd party software, no registry and the likes. The end. I can encrypt the system volume the first thing after I'm able to run a first program on the system. All the other software, all modifications to the registry and so on, will run on the encrypted disk.
Sure, of course. All you said.

Re:Encrypt it.... by h4rr4r · 2011-02-17 09:06 · Score: 1

Because they use it too? Because they would rather no one have the info than everyone?

The NSA does both securing and attacking.

Re:Encrypt it.... by smelch · 2011-02-17 09:07 · Score: 1

When did you think to yourself "Instead of worrying about the NSA decrypting my data, I should probably worry about not attracting the attention of the NSA"? I mean, if they're at the point of their investigation being stopped by your encrypted drive, are they likely to just say "Well shit, sorry about that!" and head out the door? Probably never because you're one of a billion idiots that thinks they can have and require perfect security. I bet you love making fun of the TSA too, not realizing you're making the same mistakes as them.

--
If I can just reach out with my words and touch a butthole, just one, it will all be worth it.

"Multiple times" is an exaggeration by lullabud · 2011-02-17 09:11 · Score: 1

The great zero challenge was never accepted, so I'd say it's safe to say that spinning hard disk data can reliably erased. I've never seen it done, that's for sure.

http://hardware.slashdot.org/story/08/09/06/189248/The-Great-Zero-Challenge-Remains-Unaccepted

Re:"Multiple times" is an exaggeration by Rashkae · 2011-02-17 09:27 · Score: 1

That challenge doesn't even try to prove anything. I don't want to perpetuate the myth that you need anything fancier than a single pass of 0's to wipe modern drive's before disposing on e-bay, but any theory about recovery data from a wiped drive, real or paranoid, involves removing the platter to scan it directly. The so called challenge was to recover data without dismantling the drive, so it was a dead end to begin with.
Re:"Multiple times" is an exaggeration by gstoddart · 2011-02-17 09:36 · Score: 1

The great zero challenge was never accepted, so I'd say it's safe to say that spinning hard disk data can reliably erased.
*shrug* Not my area of expertise, but in a past life the company I worked for had a product which was certified to do the secure wipe. No idea if it was no longer necessary or not. But it was done.

I've never seen it done, that's for sure.
I've never seen Platypus mate, but I'm fairly confident it happens.
Heck, the tinfoil hat crowd might say that the lettered agencies who can do this simply didn't participate.

--
Lost at C:>. Found at C.

Looks like they de-soldered by Rob+the+Bold · 2011-02-17 09:11 · Score: 1

Well I finally did read TFP referred to by the abstract in TFA mentioned in TFS. And it sure looks like they just de-soldered the ICs and popped them into a dead-bug socked on their "Ming the Merciless" custom controller board.

--
I am not a crackpot.

Most people's data is unsafe by thetoadwarrior · 2011-02-17 09:14 · Score: 1

Whether it's an hd, ssd or optical disc only a few people really care enough to secure their data and in the end if you want to make sure no one gets it the physically destroy the media when you're done. It's the safest way for all of them.

Magnets by Adam+Appel · 2011-02-17 09:15 · Score: 1

Do rare earth magnets work on SSD? Or does magnatise and destroy no longer work on today's tech?

--
They come in the dark, only in the darkest.

Re:Magnets by CookieForYou · 2011-02-17 12:49 · Score: 1

Magnets do nothing to SSDs.

So what it NSA can read it? by perpenso · 2011-02-17 09:16 · Score: 1

Someone once told me that I should use RSA encryption because it was developed by the NSA. I thought to myself "why would the NSA produce and give away an encryption algorithm they can't break". I concluded that they wouldn't. So yeah, probably not secure.

For the sake of argument lets assume the NSA can break it. So what? The government already has my SSN, bank account numbers and credit card numbers. I only need to stop the thieves, finder keepers, dumpster divers, computer recyclers, etc.

Re:So what it NSA can read it? by cheekyjohnson · 2011-02-17 10:23 · Score: 1

So what?
You may not care, but other people do.

The government already has my SSN, bank account numbers and credit card numbers.
If the situation is bad, you shouldn't make it worse. Some people merely like being secure.

--
Filthy, filthy copyrapists!
Re:So what it NSA can read it? by X0563511 · 2011-02-18 05:13 · Score: 1

Two words: Rubber Hose
The idea being that, after a point, increasing security starts increasing your risk.

--
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
Re:So what it NSA can read it? by cheekyjohnson · 2011-02-18 08:20 · Score: 1

How does that apply to this situation?

--
Filthy, filthy copyrapists!

Re:Pure crap by Rockoon · 2011-02-17 09:18 · Score: 2

Give her a gift card for a spa or other "nice" thing to do for the day. She will (A) love you for it, (B) never need to know that you had a kiddie porn drive, or (C) that you baked said kiddie porn drive in the oven while downloading midget porn as a replacement.

--
"His name was James Damore."

Hard to delete sounds like a plus ... by perpenso · 2011-02-17 09:19 · Score: 1

Encrypting it? Is taking data off really an issue anyway. If it's confidential data, destroy the disk when you need to dispose of it. Not repurposing or re-selling hardware with sensitive information on it sounds like a no-brainer.

Also if its so hard to delete then maybe SSD drives are a good place for long term backup/storage of those encrypted volumes. Just wondering, not claiming it is so.

Re:Encrypt it.... by mosb1000 · 2011-02-17 09:21 · Score: 1

US intelligence agencies clandestinely gather information. You don't lean about their investigation until you are already in custody (if at all). Who's to say they don't just automatically decrypt any RSA encoded transmission they intercept just to see what information people are trying to keep secret? They are currently involved in known projects that automatically track all telephone and internet traffic into and out of the US. Who is to say what else they're involved in?

Trusting an intelligence agency, who makes knowing your secrets its business, to encrypt your files is counter-intuitive, to say the least.

I wonder by fishbowl · 2011-02-17 09:22 · Score: 1

I wonder what the value of "remnant data" could be when the data were, say, AES encrypted?

You are encrypting your confidential data, correct? Or should I say, unencrypted data are not "confidential" in the first place?

--
-fb Everything not expressly forbidden is now mandatory.

dd? by pak9rabid · 2011-02-17 09:29 · Score: 1

Would doing a 'dd if=/dev/zero of=/dev/sda' a few times not do it?

Re:dd? by owlstead · 2011-02-17 11:14 · Score: 1

Nope. These drives contain more memory than their stated capability. Of course, with the drives getting cheaper the number of extra blocks used for wear leveling is being lowered each time. My Intel SSD contains oodles of extra memory. If you just zero the drive, you may be sure that there is a residue left. Next time try and read the article.
Re:dd? by pak9rabid · 2011-02-18 03:37 · Score: 1

Next time try and read the article.
Next time try not being a douche. It's people like you that turn this place into a clusterfuck of smug, elitist assholes.
Re:dd? by owlstead · 2011-02-19 00:43 · Score: 1

Yeah, everybody that does not read the articles get itchy. I'm fine with that. And if this means there is a lot less swearing and name calling, I can live with that too.

Amended platter removal terms by lullabud · 2011-02-17 09:36 · Score: 2

They later amended the platter removal terms with the following text, but still nobody accepted it.

If the challenger is an established data recovery business located in the United States of America (We would need to see Articles of Incorporation, a current business license and one other form of business identification in order to determine that they are indeed a professional, for-profit, established data recovery business) or a National government law enforcement or intelligence agency (NSA, CIA, FBI), then we will allow these type of organizations to disassemble the drive and to keep the drive for thirty (30) consecutive days.

They did not do whole disk wipe by losttoy · 2011-02-17 09:37 · Score: 1

AFAIK, they did not do whole disk wipe.

The website says "Individual file sanitization techniques, all of which failed and left at least 10MB of a 1000MB file." Does not say what happens when you do a full disk wipe. #Fail.

Re:Encrypt it.... by AndrewNeo · 2011-02-17 09:37 · Score: 1

So do you always just believe what people tell you without checking facts?

Re:Secure Erase by TheSpoom · 2011-02-17 09:42 · Score: 1

If TFA is correct and you can't reliably erase data from the disk, how are you going to reliably erase the key?

--
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs

Does this mean essentially infinite capacity? by mcmonkey · 2011-02-17 10:07 · Score: 1

What am I missing here? I have a drive/card/chip labeled 16 GB storage. I save 16 GB of data to it. I overwrite the entire volume with 1s.

Now I can read 16 GBs of 1s. And some l33t hacker can retrieve the 16 GB of secret sauce I thought was overwritten. So a drive labeled 16 GB really has 32 GB capacity, it's just that second 16 GB is hard to access?

And what if I then go back and overwrite those 1s with 0s or random bits? Is it possible to retrieve the layer of 1s and the original data? So a 16 GB disk can hold 48 GB? And that last 32 GB is just really, really hard to access?

Of course, I didn't RTFA. But I presume we're not just talking about delete/undelete of single files. If they didn't wipe the whole disk, why would this be on /.?

Why is this even an article? by bertok · 2011-02-17 10:10 · Score: 1

Why is Slashdot posting these inane articles?

Everybody who knows anything about SSDs knows that they have significantly more raw storage than logical capacity, and that the extra storage capacity is used for redundancy. Because of the wear levelling systems used, writes don't go back to the same place, so data can't be overwritten. This has been well known and obvious to everyone for years.

Pro Tip: Full Disk Encryption. Problem. Fucking. Solved.

Why are we even talking about this?

Re:Why is this even an article? by geekoid · 2011-02-17 11:02 · Score: 1

You are a myopic fool.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

Use the Hollywood Method to Destroy It by Maltheus · 2011-02-17 10:15 · Score: 1

Just mount the flash drive and put a bullet in your computer's monitor. If it works for hard drives in the movies, then it should work for solid state as well.

Re:Because things are really analog not digital .. by CookieForYou · 2011-02-17 10:15 · Score: 1

As it has been pointed out, modern drives overlap various bits quite a bit and there really is no such residual magnetism. It is below the noise floor of the natural variations in a platter's magnetism.

Some of this research is even from the same guy (Guttmann) who published the technique 25 years ago, but states it is impossible with modern drives.

Re:Encrypt it.... by fishbowl · 2011-02-17 10:26 · Score: 1

That in itself would be a bigger secret than anything that would be exposed by it.
Acting on any information obtained that way would expose the biggest secret in the history of secrecy, pretty much.
They aren't going to tip that hand just for you.

--
-fb Everything not expressly forbidden is now mandatory.

Re:Because things are really analog not digital .. by vadim_t · 2011-02-17 10:27 · Score: 1

And even if you can read this residual magnetism, think of what you must do next:

First, a drive head isn't enough. You have to get the platter under an electron microscope or such incredibly specialized device owned by what, 10 labs in the whole world?

Next, you spend months (from what I heard of the speed you get out of those) copying the platter, generating several times more data than the official disk's capacity.

Once you're done with that you can get to decoding. But, there's a laboratory proof of concept, and there's the real thing. On a real drive, you won't get a laboratory setting of showing you can read sector #1 and then figure out what the previous value was. You'll have to find something interesting in millions of sectors.

On hard disks data doesn't get written in neat tidy ways. Files get fragmented all over the platter, and when deleted their sectors may get reused. So you'll have to find your interesting file by piecing it together. You'll have to make sense of the former filesystem metadata that says where it was, then read the now overwritten file. Both of which are probably not neatly overwritten once, but a different amounts of times on each sector, and you'll have to figure out which of those is the good one.

It sounds like way, way too much trouble.

Not terribly different from a HDD by izomiac · 2011-02-17 10:28 · Score: 1

What's an even bigger concern is that when an SSD fails, your whole disk is still available read-only. I've got one sitting around like that, and have been too lazy to physically destroy it (none of the data is sensitive). What I should have done is just turn it into super-fast installation media for a few versions of Windows, but I wasn't thinking at the time.

Of course, that failure model is a *feature* of SSDs. With a HDD, the drive just randomly fails someday, and you lose the ability to read, write, or securely erase data. If you have sensitive data, it shouldn't be stored on any media unless it's encrypted or physically/remotely secure and will be throughly destroyed when it dies. That's common sense. Blocks being difficult to securely erase due to wear leveling and such doesn't change that.

Czo... who? by mangu · 2011-02-17 10:30 · Score: 1

It's easy to get the data off

Much easier than spelling the inventor's name

Well duh! by wcrowe · 2011-02-17 10:39 · Score: 1

Of course they can always get the data off. Everyone knows that. They do it all the time on CSI. Sheesh!

--
Proverbs 21:19

Condoms are similar. by shihonage · 2011-02-17 10:50 · Score: 1

Once you put it on, getting off isn't so easy.

Re:Encrypt it.... by mosb1000 · 2011-02-17 10:51 · Score: 1

So what you do is you use that information to lead you to other evidence you can use. During WW2, they would send airplanes to verify submarine locations before destroying them so that the Germans wouldn't get wise to them. Presumably they've had 70 years or so to improve on that technique. I think it's safe to say they can figure it out.

Re:Just use a more powerful data descrution device by geekoid · 2011-02-17 11:04 · Score: 1

Yes, that's what In tell security when walk around our server room with fire arms.

Or maybe shooting doesn't actually solve all your problems.

--
The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

Re:Just use a more powerful data descrution device by h4rr4r · 2011-02-17 11:09 · Score: 1

You take the drives to the range, not go shooting in the server room.

Re:Secure Erase by owlstead · 2011-02-17 11:09 · Score: 1

No, that would be highly unlikely (that it is encrypted by default). You do need to set some security parameters when erasing, but as far as I know, that's not because of how secure erase works. The results in the article for secure erase would not be possible if there was a single key (because partial erase would not be possible). It also would not explain the 20 second wait during secure erase of my Intel SSD. Fortunately, flash blocks can be erased in one go, so secure erase is *much much much* faster on an SSD compared to a HDD.

Re:Secure Erase by owlstead · 2011-02-17 11:11 · Score: 1

That would be possible, since it would be the device itself that does the erasing. It could just issue a single erase command for the flash blocks containing the key and it would be done (I presume that the key is stored in at least two locations, or the failure of a single block of memory would be disastrous). It might even be stored in writeable memory within the controller.

Crush the competition! by mbstone · 2011-02-17 11:17 · Score: 1

1. Buy a steamroller.
2. Get government contract for SSD data destruction.
3. Profit!

Rsa is 3 guys by Nicolas+MONNET · 2011-02-17 12:10 · Score: 1

And it's not the NSA.

Re:Encrypt it.... by icebraining · 2011-02-17 12:58 · Score: 1

RSA was not developed by the NSA, but by Rivest, Shamir and Adleman at MIT.

--
Dilbert RSS feed

Suggested Alternative Headline by severoon · 2011-02-17 13:33 · Score: 1

Suggested Alternative Headline: "Confidential, Unencrypted Data Not Safe on Solid State Disk, Conventional Disk, or Anywhere Else Now That I Think About It"

--
but have you considered the following argument: shut up.

Re:Encrypt it.... by Thing+1 · 2011-02-17 14:21 · Score: 1

When did you think to yourself "Instead of worrying about the NSA[...]" [...] I bet you love making fun of the TSA too [...]

Okay, let's get all the future agencies out of the way; we've already got NSA and TSA. Feel free to participate.

ASA
BSA (damn scouts/Microsoft!)
CSA
DSA
ESA (Euroooooos iiiiin spaaaace!)
FSA (use it or lose it)
GSA
HSA (use this or lose it also)
ISA (old motherboards)
JSA
KSA (Kal-el space agency?)
LSA (heh, woodrose/glory)
MSA
[...]
OSA
PSA (walnut-sized; stimulate it by going backwards)
QSA
RSA (rot-26)
SSA (you won't get the payout)
[...]
USA (yeah right)
VSA
WSA
XSA
YSA
ZSA (slap a fucking policeman!)

--
I feel fantastic, and I'm still alive.

Re:Encrypt it.... by Thing+1 · 2011-02-17 14:25 · Score: 1

That in itself would be a bigger secret than anything that would be exposed by it. Acting on any information obtained that way would expose the biggest secret in the history of secrecy, pretty much. They aren't going to tip that hand just for you.

Exactly: similar to the reason we let that town be destroyed in England; alerting them would have let the Germans know we figured out the Enigma, and those lives were deemed to be less important than "the war effort". (Note to self: stay out of the way of war efforts.)

--
I feel fantastic, and I'm still alive.

Mechanical Disks Aren't Entirely Different by bill_mcgonigle · 2011-02-17 15:57 · Score: 1

I know OCZ has its own wipe utility and I believe intel too. Using wiping software designed for mechanical disks makes absolutely no sense and the results from this study are 100% predictable. Oh your Gutmann wipe pattern for circa1991 MFM drives doesn't wipe SSDs? You don't say! If you needed to securely wipe one, use the proper tool.

Even mechanical disks need this - if you get a sector re-mapped, you're not going to zero it out ever again.

Some SATA drives support a Secure Erase ATA command extension. I asked Seagate to send me a list of their drives that had this support in firmware, so I could write a tool to do this. They refused. Even as a "Seagate Partner".

So, in the general case, you can't trust your drives. LUKS is easy enough to set up on Linux that you can work around drive vendors you can't trust (but set swappiness to 0 on a netbook!).

--
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)

Re:Encrypt it.... by The+Wild+Norseman · 2011-02-17 17:53 · Score: 1

I bet you love making fun of the TSA too, not realizing you're making the same mistakes as them.

No, I don't let anyone get a free pass. I scan everyone that comes into my house with an old x-ray machine I found once while dumpster diving behind some doctor's clinic. I figure that a little extra radiation won't hurt nobody and fuck 'em since they agreed to come into my house when I called them to fix my sink.

--
"A government is a body of people usually -- notably -- ungoverned." -Shepherd Book

Re:Encrypt it.... by Neil+Boekend · 2011-02-17 19:06 · Score: 1

Encrypt it with quad rot-13. They will not be able to crack that!

--
Well, I might have a way, but it only works on a semi spherical planet in a vacuum.

When You Absolutely Need Something Destroyed..... by IHC+Navistar · 2011-02-17 19:13 · Score: 1

Call the Marines.

--
Knowing Google's lust for data collection, the Soviet Union is still alive and well inside the psyche of Sergey Brin....

Full disk encryption by 1s44c · 2011-02-17 21:42 · Score: 1

It eats some of the speed advantages but whole disk encryption works for me. It's still way faster than a magnetic disk.

dd is not enough. by leuk_he · 2011-02-17 22:55 · Score: 1

No it is not.

dd was was fine in the year 2000.
It does not work today for the following reasons:
Harddrive do re-maps for bad blocks. These bad blocks are not touched by OS tools.
SSD does this even more aggressive and even by default keeps a pool (10%) of flash just to recover form material defects and might alos compress data (e.g. write all zero bytes and it will compress the data) to minimize the number of writes.

In theory the security erase tools send the disk a low level command that will really zero all data, but the investigators did show that this optional STA command was not implemented correctly in some cases.

dd if =/dev/urandom of=/dev/sdxxx will probably erase the data, BUT NOT ALL OF THE DATA ON A SSD, rewriting with zero's might be a non-productieve idea with advanced disk firmwares.

PS, I agree that overwriting the data multiple times that some old tools did is just a waste of time, on a SSD it will only cause more wear.

Probably already said but... by beaverdownunder · 2011-02-18 01:01 · Score: 1

the old tried-and-true method of 'securing' your data still applies: woodchipper.

Re:Secure Erase by TheSpoom · 2011-02-18 01:21 · Score: 1

If it's stored in a completely different type of memory, the always-crypto delete-the-key approach might work, but if it's actually on the SSD itself, it falls apart. You said yourself that the key is likely stored in two locations in case one fails; in that case, wouldn't at least part of the key be recoverable on the fails portion of the disk since it may refuse to write to that block, since it considers it damaged?

--
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs

overwritten data remaining intact by doperative · 2011-02-18 03:11 · Score: 1

Is it possible to recover data after running the following a number of times?

$dd if=/dev/urandom of=/dev/hda

Re:overwritten data remaining intact by vuffi_raa · 2011-02-18 09:44 · Score: 1

that's what I was wondering myself
Re:overwritten data remaining intact by vuffi_raa · 2011-02-18 11:21 · Score: 1

I just tried it on a usb drive and I am not finding any data, the article seems to indicate that shredding programs ignore some sort of extra flash space on it, I am curious to know if that is the case because a $ sudo dd if=/dev/sdb | hexdump -C | head shows nothing but zeros.
do you have to take the drive apart to access retreivable data?
Re:overwritten data remaining intact by PipsqueakOnAP133 · 2011-02-23 20:33 · Score: 1

Yes. Yes you do.
You won't necessarily get your data back, but there's a chance you can find something in there.

Condifentiality of data not safe ... by mlinksva · 2011-02-18 06:33 · Score: 1

Is what the story title should've been. Confidentiality, not data, is the subject of "safe". Much like copyright doesn't "protect" creative works, rather it protects revenue streams and feelings of copyright holders and authors.

If you are really serious.... by joerog · 2011-02-19 03:22 · Score: 1

To truly delete or protect the info on the SSD, after copying all the pertinent data, remove the SSD from the slot, place it on a solid object (such as a brick) and administer a strong concussive blow with a nail-driving device (ie Hammer). The SSDs are cheap enough and if your data is as valuable as you think it is, there is no great loss (of the SSD). If you think this is a waste, then perhaps your data is really not that valuable after all......

Slashdot Mirror

Confidential Data Not Safe On Solid State Disks

299 of 376 comments (clear)