What shredding utility can I use?
I happen to know that formatting a disk even 10 times does not destroy everything.
Is there a good shredding utility I can use on Linux ?
formatting file-shredding
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
add a comment |
I happen to know that formatting a disk even 10 times does not destroy everything.
Is there a good shredding utility I can use on Linux ?
formatting file-shredding
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
2
Zero-filling does.ddis enough for that.
– Dennis
Jan 26 '13 at 0:16
Need to do it more than once. And even so data can be recovered. Useshred(1)on the disk. It also works on files, but not on COW-based filesystems like btrfs.
– vonbrand
Jan 26 '13 at 7:56
1
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44
add a comment |
I happen to know that formatting a disk even 10 times does not destroy everything.
Is there a good shredding utility I can use on Linux ?
formatting file-shredding
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
I happen to know that formatting a disk even 10 times does not destroy everything.
Is there a good shredding utility I can use on Linux ?
formatting file-shredding
formatting file-shredding
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
edited Jan 26 '13 at 1:12
Joseph Quinsey
4671723
4671723
asked Jan 25 '13 at 23:58
statquantstatquant
13515
asked Jan 25 '13 at 23:58
statquantstatquant
13515
asked Jan 25 '13 at 23:58
statquantstatquant
13515
13515
2
Zero-filling does.ddis enough for that.
– Dennis
Jan 26 '13 at 0:16
Need to do it more than once. And even so data can be recovered. Useshred(1)on the disk. It also works on files, but not on COW-based filesystems like btrfs.
– vonbrand
Jan 26 '13 at 7:56
1
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44
add a comment |
2
Zero-filling does.ddis enough for that.
– Dennis
Jan 26 '13 at 0:16
Need to do it more than once. And even so data can be recovered. Useshred(1)on the disk. It also works on files, but not on COW-based filesystems like btrfs.
– vonbrand
Jan 26 '13 at 7:56
1
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44
2
2
Zero-filling does.
dd is enough for that.– Dennis
Jan 26 '13 at 0:16
Zero-filling does.
dd is enough for that.– Dennis
Jan 26 '13 at 0:16
Need to do it more than once. And even so data can be recovered. Use
shred(1) on the disk. It also works on files, but not on COW-based filesystems like btrfs.– vonbrand
Jan 26 '13 at 7:56
Need to do it more than once. And even so data can be recovered. Use
shred(1) on the disk. It also works on files, but not on COW-based filesystems like btrfs.– vonbrand
Jan 26 '13 at 7:56
1
1
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44
add a comment |
3 Answers
3
active
oldest
votes
With conventional hard drives, a single wipe with zeros may be enough
The 'multiple wipes' method assumes that you're using older drives (with larger magnetic domains). The 'definitive' paper on data destruction by Guttmann suggests 35 different patterns - which are effective on different types of drives.
Guttmann suggests filling the drive with random data these days and a single wipe would do according to most. The shred command works for that.
The situation with SSDs is more muddied. Bell and Boddington at Murdoch University claim that the garbage collection on SSDs tends to overwrite deleted data in their paper. A team at the University of California claims the exact opposite, that nothing short of physical destruction works and that both ATA secure delete and shredding methods fail in most cases. Taking all this into account toolwise, you should consider shred (which does a high level secure wipe), and running a SATA secure wipe from HDparm if you can which is at lower level. That should handle most situations I believe. You should also consider encrypting any data worth deleting from the get go.
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,shreddoes nothing thatddcan't do, and neither are effective for wiping flash drives.
– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
add a comment |
For entire disks, there's the shred command, which by default only overwrites three times but with the -n <number> option can do as many passes as desired. It doesn't work so well on individual files in journaled filesystems, though. With the -z option, shred will do an extra pass with 0s afterward, so the shredding isn't immediately obvious.
add a comment |
You can copy rubbish multiple times.
for i in `seq 1 35`; do
dd if=/dev/urandom of=/dev/sdX
done
answered Jan 26 '13 at 0:01
ssicessice
673512
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,/dev/urandomtakes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).
– Dennis
Jan 26 '13 at 3:03
It's got no advantage overshred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.
– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
|
show 3 more comments
protected by Ramhound 2 days ago
Thank you for your interest in this question.
Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).
Would you like to answer one of these unanswered questions instead?
3 Answers
3
active
oldest
votes
3 Answers
3
active
oldest
votes
active
oldest
votes
active
oldest
votes
With conventional hard drives, a single wipe with zeros may be enough
The 'multiple wipes' method assumes that you're using older drives (with larger magnetic domains). The 'definitive' paper on data destruction by Guttmann suggests 35 different patterns - which are effective on different types of drives.
Guttmann suggests filling the drive with random data these days and a single wipe would do according to most. The shred command works for that.
The situation with SSDs is more muddied. Bell and Boddington at Murdoch University claim that the garbage collection on SSDs tends to overwrite deleted data in their paper. A team at the University of California claims the exact opposite, that nothing short of physical destruction works and that both ATA secure delete and shredding methods fail in most cases. Taking all this into account toolwise, you should consider shred (which does a high level secure wipe), and running a SATA secure wipe from HDparm if you can which is at lower level. That should handle most situations I believe. You should also consider encrypting any data worth deleting from the get go.
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,shreddoes nothing thatddcan't do, and neither are effective for wiping flash drives.
– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
add a comment |
With conventional hard drives, a single wipe with zeros may be enough
The 'multiple wipes' method assumes that you're using older drives (with larger magnetic domains). The 'definitive' paper on data destruction by Guttmann suggests 35 different patterns - which are effective on different types of drives.
Guttmann suggests filling the drive with random data these days and a single wipe would do according to most. The shred command works for that.
The situation with SSDs is more muddied. Bell and Boddington at Murdoch University claim that the garbage collection on SSDs tends to overwrite deleted data in their paper. A team at the University of California claims the exact opposite, that nothing short of physical destruction works and that both ATA secure delete and shredding methods fail in most cases. Taking all this into account toolwise, you should consider shred (which does a high level secure wipe), and running a SATA secure wipe from HDparm if you can which is at lower level. That should handle most situations I believe. You should also consider encrypting any data worth deleting from the get go.
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,shreddoes nothing thatddcan't do, and neither are effective for wiping flash drives.
– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
add a comment |
With conventional hard drives, a single wipe with zeros may be enough
The 'multiple wipes' method assumes that you're using older drives (with larger magnetic domains). The 'definitive' paper on data destruction by Guttmann suggests 35 different patterns - which are effective on different types of drives.
Guttmann suggests filling the drive with random data these days and a single wipe would do according to most. The shred command works for that.
The situation with SSDs is more muddied. Bell and Boddington at Murdoch University claim that the garbage collection on SSDs tends to overwrite deleted data in their paper. A team at the University of California claims the exact opposite, that nothing short of physical destruction works and that both ATA secure delete and shredding methods fail in most cases. Taking all this into account toolwise, you should consider shred (which does a high level secure wipe), and running a SATA secure wipe from HDparm if you can which is at lower level. That should handle most situations I believe. You should also consider encrypting any data worth deleting from the get go.
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
With conventional hard drives, a single wipe with zeros may be enough
The 'multiple wipes' method assumes that you're using older drives (with larger magnetic domains). The 'definitive' paper on data destruction by Guttmann suggests 35 different patterns - which are effective on different types of drives.
Guttmann suggests filling the drive with random data these days and a single wipe would do according to most. The shred command works for that.
The situation with SSDs is more muddied. Bell and Boddington at Murdoch University claim that the garbage collection on SSDs tends to overwrite deleted data in their paper. A team at the University of California claims the exact opposite, that nothing short of physical destruction works and that both ATA secure delete and shredding methods fail in most cases. Taking all this into account toolwise, you should consider shred (which does a high level secure wipe), and running a SATA secure wipe from HDparm if you can which is at lower level. That should handle most situations I believe. You should also consider encrypting any data worth deleting from the get go.
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
edited Nov 5 '15 at 4:52
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
answered Jan 26 '13 at 1:27
Journeyman Geek♦Journeyman Geek
112k43216366
112k43216366
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,shreddoes nothing thatddcan't do, and neither are effective for wiping flash drives.
– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
add a comment |
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,shreddoes nothing thatddcan't do, and neither are effective for wiping flash drives.
– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
Hi, thanks, I had a presentation from our departement of defense on computer security, I was amazed by the lack of security and how easy it was to hack un protected systems.
– statquant
Jan 26 '13 at 9:08
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,
shred does nothing that dd can't do, and neither are effective for wiping flash drives.– Hashim
Dec 24 '18 at 18:58
This answers cites the Usenix paper on the faultiness of Secure Erase implementations and then goes on to recommend using Secure Erase. Further,
shred does nothing that dd can't do, and neither are effective for wiping flash drives.– Hashim
Dec 24 '18 at 18:58
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
Gutman's paper has been widely misinterpreted. Gutman never actually said that those 35 patterns were necessary, only that he thought they were likely sufficient, given his speculations about how data retrieval might be possible - but he never cited any examples of it ever having been done. And there are sound reasons for believing that a single overwrite with random data is sufficient to wipe any hard drive of any technology.
– Jamie Hanrahan
2 days ago
add a comment |
For entire disks, there's the shred command, which by default only overwrites three times but with the -n <number> option can do as many passes as desired. It doesn't work so well on individual files in journaled filesystems, though. With the -z option, shred will do an extra pass with 0s afterward, so the shredding isn't immediately obvious.
add a comment |
For entire disks, there's the shred command, which by default only overwrites three times but with the -n <number> option can do as many passes as desired. It doesn't work so well on individual files in journaled filesystems, though. With the -z option, shred will do an extra pass with 0s afterward, so the shredding isn't immediately obvious.
add a comment |
For entire disks, there's the shred command, which by default only overwrites three times but with the -n <number> option can do as many passes as desired. It doesn't work so well on individual files in journaled filesystems, though. With the -z option, shred will do an extra pass with 0s afterward, so the shredding isn't immediately obvious.
For entire disks, there's the shred command, which by default only overwrites three times but with the -n <number> option can do as many passes as desired. It doesn't work so well on individual files in journaled filesystems, though. With the -z option, shred will do an extra pass with 0s afterward, so the shredding isn't immediately obvious.
answered Jan 26 '13 at 0:04
community wiki
Darael
add a comment |
add a comment |
You can copy rubbish multiple times.
for i in `seq 1 35`; do
dd if=/dev/urandom of=/dev/sdX
done
answered Jan 26 '13 at 0:01
ssicessice
673512
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,/dev/urandomtakes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).
– Dennis
Jan 26 '13 at 3:03
It's got no advantage overshred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.
– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
|
show 3 more comments
You can copy rubbish multiple times.
for i in `seq 1 35`; do
dd if=/dev/urandom of=/dev/sdX
done
answered Jan 26 '13 at 0:01
ssicessice
673512
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,/dev/urandomtakes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).
– Dennis
Jan 26 '13 at 3:03
It's got no advantage overshred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.
– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
|
show 3 more comments
You can copy rubbish multiple times.
for i in `seq 1 35`; do
dd if=/dev/urandom of=/dev/sdX
done
answered Jan 26 '13 at 0:01
ssicessice
673512
You can copy rubbish multiple times.
for i in `seq 1 35`; do
dd if=/dev/urandom of=/dev/sdX
done
answered Jan 26 '13 at 0:01
ssicessice
673512
answered Jan 26 '13 at 0:01
ssicessice
673512
answered Jan 26 '13 at 0:01
ssicessice
673512
answered Jan 26 '13 at 0:01
ssicessice
673512
673512
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,/dev/urandomtakes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).
– Dennis
Jan 26 '13 at 3:03
It's got no advantage overshred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.
– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
|
show 3 more comments
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,/dev/urandomtakes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).
– Dennis
Jan 26 '13 at 3:03
It's got no advantage overshred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.
– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
That unnecessarily slow. See my answer to zero fill vs random fill.
– Dennis
Jan 26 '13 at 1:12
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
@Dennis it's not that slower than zeroing the disk, and it provides a higher security level. If you are zeroing a disk, maybe 1 minute more or less won't really matter.
– ssice
Jan 26 '13 at 2:44
It is much, much slower. On my machine,
/dev/urandom takes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).– Dennis
Jan 26 '13 at 3:03
It is much, much slower. On my machine,
/dev/urandom takes 52 seconds to produce 1 GB of output. That's more than 14 hours for 1 TB, and 21 days for overwriting the disk 35 times. In comparison, zeroing a 1 TB hard drive once should take less than 3 hours (assuming 100 MB/s avg. write speed).– Dennis
Jan 26 '13 at 3:03
It's got no advantage over
shred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.– Darael
Jan 26 '13 at 18:10
It's got no advantage over
shred, though (both will deplete the entropy pool to some extent, this being the downside of random writes), and it's rather a lot more to type.– Darael
Jan 26 '13 at 18:10
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
@Darael Sure. Your answer and mine were more or less simultaneous, I didn't see yours. However, I don't think there's a need to downvote.
– ssice
Jan 26 '13 at 18:13
|
show 3 more comments
protected by Ramhound 2 days ago
Thank you for your interest in this question.
Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).
Would you like to answer one of these unanswered questions instead?
2
Zero-filling does.
ddis enough for that.– Dennis
Jan 26 '13 at 0:16
Need to do it more than once. And even so data can be recovered. Use
shred(1)on the disk. It also works on files, but not on COW-based filesystems like btrfs.– vonbrand
Jan 26 '13 at 7:56
1
I always scratch my head over posts like these. Not to many people have the equipment to read data that has been overwritten even once. And those that do are likely to get your data before you can erase it...
– Keltari
Jan 26 '13 at 9:27
@Keltari: I hear you, but still it is good to know
– statquant
Jan 26 '13 at 12:44