Is it really enough to overwrite my disk with one pass zeroes?
Is then everything gone like file names, dates, type etc
>>52087345
Yes
>>52087345
For what purpose?
Resale?
Throwing it away?
If you're really iffy about this, just swipe a magnet over the drive.
>>52087345
One pass is plenty. Do two if you're really paranoid.
Smash drive and throw into 10 different trash bins, one in each state, if spase people are after ur databits
>>52087345
no, one pass zeroes is stupid. do one pass random data.
No goy you need at least 35 pass Gutmann followed by swiping a magnet over your dick 1,000,000 time.
The latent magnetic history of the bits is readable for several cycles. You should run at least 25 random passes (making sure the random number generator is legit made for this specific purpose) otherwise people can still lift your data.
>>52089142
source?
No.
Drives bit space has extra memory for "caching" of historical bits. This is what NSA/gov agency uses to decode deleted files.
You need to over write that cache by performing dban multiple times. If you only do it 1-2 the cache will still remember what the data was 1-2 writes ago. I'm not sure what the exact limit is but there is a limit to it and most only use 2-3 recent bit memory recovery technique. So its good to do it at least 3 times.
Note: This technology might have advanced further with SSD/modern drives and thus allowing longer storage of bit history. So it might be bit outdated.
>>52087345
you can tell if that zero was a 1 or 0, that's why you need more than ~5 pass in random, is not like a haxx0r could figure it out it in the house tho
>>52087729
maybe he's a pedo. haha!
>>52087345
just empty your recycle bin after you delete sensitive data and you'll be fine.
trust me I've lurked here for years.
>>52089282
bullshit detected
otherwise, source pls
>>52089250
What for? It is known.
>>52089632
hello neo /g/
shred -vzu --random-source=/dev/urandom -n 8 /dev/sd*
>>52087345
Yes
>>52090785
>It is known.
Post a single real world example as your source then.
If you're just tossing it out or selling it then one pass with zeros is enough.
>>52090755
Bit of research is needed, but its there. Look under "bit cell redundancy" or "hdd error correction" to get a glimpse of this. This is also done with SSD/flash storage as well.
That is how its normally presented as, but the secondary function of this is to allow persons with certain keys to get those data manually/forcibly. Its not a hard thing to imagine.
If you're selling your pc/laptop/hdd/ssd, and you didn't have any very important data on it, just overwrite the data twice, if you're really worried 5 passes should get rid of everything. And just make sure you don't sell it to someone who would want the data, and would do their best to recover whatever is still recoverable. And it's best to use something that will overwrite it randomly and not only with 0. If it's a pc/laptop you could just put a new hdd in it and do a fresh install of the os.
>>52091179
Bruce Schneier said that using one pass makes it impossible for anyone to recover shit
>>52091310
That was during the HDD era. Probably the early HDD era. But again, with new technology I'd do at least 2-3 times.
>>52091345
No, its literally the opposite of what you are saying.
Old HDD's needed more passes because of the density they had. New ones need only one pass (excluding SSD).
>>52091179
>still no source
There's no example of a readable file being recovered from a zeroed drive, anywhere, ever.
>>52091399
Take it as you will. The technology isn't unfeasible and its already implemented in most storage disks.
>>52091391
Old HDD had shit density this also means manufacturers couldn't reliably do ecc efficiently without sacrificing performance/storage capacity. Most people from the 90s would say you need multiple writes to erase all the sectors, but many still said 1 pass would be enough back then as well.
Modern HDD/SSD allows much more efficient way to store multiple bits per cell (old days, you could only store 1 bit per cell). Efficiency, miniaturization and accuracy allows much more effective ecc without sacrificing performance or capacity storage.
>>52091521
Yes it is unfeasible. That's why there's no example of a readable file being recovered.
A random bit here and there, sure. Since you need 8 bits for a single byte that's hardly useful if you're trying to recover any meaningful data that's actually readable.
It's like me giving you 3 digits of my 16 digit credit card number at random. Beyond being able to say that, technically, you know part of my credit card number, it's completely useless information.
>>52088397
don't do random. it looks too much like the disk is encrypted. authorities will pressure/torture you for a non-existent password
>>52091849
verses what, a drive of nothing but zeroes?
why not just edit the data in the file followed by encryption then open the file in a editor (edit/notepad/whatever you prefer) add garbage data and delete random bits then delete the file
would that be enough?
>>52091877
all zeroes make it look crescent fresh
>>52091849
lol
>>52087345
Do a random data instead. It is more effective and will take the same amount of time.
>>52091893
a bit can only be a 0 or a 1. a single pass of just 0s might leave enough residual magnetism of the bits that were 1s to recover data.
best way is do a pass of all 1s and then a pass of all 0s. that way the drive looks like it hasn't been used and it will be very difficult to tell any bit's original value.
I use full disk encryption. And then zero it.
>>52087345
One Zeroes Pass? NO.
One Random Pass? Yes.
