0

I use cat /dev/zero to write on free space on Each drive one by one but my question should I use;

cat /dev/zero > wipe_space or cat /dev/urandom > wipe_space

2.I know that first write zero and the second write random data, but zeros is much faster, so whats the difference?

3.and Is there any difference between cat , dd command?

4.last question is this way secure just dont want the one who will get my pc restore any thing with recovery tools.

Note:I'll not erase every thing, 'll delete personal stuff and wipe free space!

illsecure
  • 25
  • 1
  • 4
  • It is more useful to ask individual questions separately (so people can find similar problems to their own easily): http://meta.stackexchange.com/questions/39223/one-post-with-multiple-questions-or-multiple-posts – Kevin Nov 17 '12 at 08:30
  • See my answer to this question where I quote a view that writing zeros is probably safe enough for private use. – Takkat Nov 17 '12 at 09:14
  • See also https://askubuntu.com/a/1116472/295286 – Sergiy Kolodyazhnyy Feb 07 '19 at 20:28
  • Also, for securely erasing hard drives you may want to use shred command instead https://askubuntu.com/q/17640/295286 for SSD use secure erase ATA https://security.stackexchange.com/a/175972/121824 – Sergiy Kolodyazhnyy Feb 07 '19 at 20:29

1 Answers1

0

  1. I would imagine /dev/zero would be quicker. You could test by using the test command in the terminal.

  2. Advanced recovery tools can analyse the residual magnetic fields left on the disk blocks in order to recover data that has been overwritten. The safest way completely destroy data is to write random data to the disk multiple times (this all assumes you are using a magnetic disk, and not an SSD).

Kevin
  • 380
  • why write random data to the disk multiple times, Is the normal use can recover one time written files with his personal computer? and why zero quicker than urandom? – illsecure Nov 17 '12 at 08:42
  • In all likeliness, writing zeros once will be enough to hide the data. Writing all zeros will leave traces of the original data, similar to using an eraser to erase pencil. More info: http://en.wikipedia.org/wiki/Data_remanence – Kevin Nov 17 '12 at 08:48
  • Recovering data from residual traces requires a professional laboratory and specialized drive firmware (and thus a lot of money). Yes it is possible, but unless someone is absolutely convinced of finding very valueable data, it will not be done. – jippie Nov 17 '12 at 10:25
  • 1
    "See this answer." -> Your link is dead, this is also why you should always describe roughly what you're linking to, I think. – Christian Feb 07 '19 at 12:08
  • Noted! I'll do so in future. Removed the broken link. – Kevin Feb 07 '19 at 20:24