You are not logged in.
Has anyone ever "shredded" a hard drive before?
This is what I have... after 9 hours!
shred: /dev/sdb: pass 4/26 (ffffff)...66GiB/150GiB 44%
Only 4 of 26? How long is this supposed to take? If I killed it now would the drive be wiped from the first pass (first pass used (random) in place of ffffff, btw), or do I need to let it finish to get a full and proper shred? This is a 160GB/7200 IDE drive if that's important. Letting it run is not a big deal, I just didn't expect it to take this long.
Offline
A long time.
Archi686 User | Old Screenshots | Old .Configs
Vi veri universum vivus vici.
Offline
A long time.
I gathered.
It will take a couple of days at this rate -- unless maybe it speeds up more as it progresses? I have someone who wants to buy the drive, and I just wondered if the data is safely wiped by now, or how long it's likely to take?
Offline
Offline
great - thanks for the link.
Offline
I have seen a much better explanation quite some time ago so I don't even know how to find it. The short answer is one pass should do just fine. The thing is that in modern disks the way the data is recorded to disk and the density is so high that when you overwrite it it's for good. Reading the data back is more of a statistical process rather than a deterministic one as it used to be with old disks.
There is also another way to erase a disk, there is a smart command that does that for you (safe erase) I believe, but I don't know how long it takes.
R00KIE
Tm90aGluZyB0byBzZWUgaGVyZSwgbW92ZSBhbG9uZy4K
Offline
Wintervenom wrote:great - thanks for the link.
From that link you need 3 wipes
"One up, one down, one to polish."
(from Red Dwarf)
Last edited by ploub (2009-02-09 13:22:54)
Offline
userlander wrote:Wintervenom wrote:great - thanks for the link.
From that link you need 3 wipes
"One up, one down, one to polish."
(from Red Dwarf)
I restarted it last night with 4 passes, 3 data and then a pass of zeros, this time after formatting the drive ext2. I thought that would be faster because of having no journal (man page also tells the limitations of using it on a journaled fs) but if anything it's been even slower. 12+ hours later and it's still only halfway through, on pass 2 of 4.
Apparently just zeroing the drive once is not enough, because with forensic techniques they can tell the difference between a zero that had a zero under it, and a zero that had a 1 under it, and use that info to recover the data. Random generation on its own is thought to be basically a joke - I guess it's easier to recover than even with all zeros, because obviously some of the 1s and 0s are going to be overwritten with the same number that they were before, i.e., either a 1 or a 0. Not that it's going to matter to me, I'm just selling the drive to a regular user, but I thought it was interesting to know.
In the comments section, the guy second from the bottom here had some good points:
http://linuxhelp.blogspot.com/2006/06/h … efore.html
I can't believe that example gives -n 100 as a sane option. On a big drive, that could take weeks!
Last edited by userlander (2009-02-09 14:22:01)
Offline
Lets spice this thread a little bit
http://16systems.com/zero/index.html
http://www.abkdata.com/overwritten-data.html
R00KIE
Tm90aGluZyB0byBzZWUgaGVyZSwgbW92ZSBhbG9uZy4K
Offline
I shredded the previous RAID array in my server - 3x 200 GB in RAID 5 . I think I specified 15 passes, but I accidentally cut the power after a week and decided it had been good .
Yes I know, I'm paranoid. It seems that with powerful tools you can still manipulate data after multiple shreds though.
Got Leenucks? :: Arch: Power in simplicity :: Get Counted! Registered Linux User #392717 :: Blog thingy
Offline