How long does a 7 Pass US DoD 5220 method take?
February 6, 2008 8:04 PM Subscribe
How long will it take to do a 7 pass US DoD 5220 Compliant data wipe on a 120 GB Hard Drive?
I am using TrueCrypt to encrypt my entire drive. I have made the rescue disk and am ready to start encrypting it. But there is an option to use either a 3 or 7 pass US DoD 5220 Compliant data wipe method or the "Gutmann" 35 pass method. They caution that the Gutmann method may take several weeks. But does anyone know how long the 7 pass US DoD 5220 will take? Thanks!
I am using TrueCrypt to encrypt my entire drive. I have made the rescue disk and am ready to start encrypting it. But there is an option to use either a 3 or 7 pass US DoD 5220 Compliant data wipe method or the "Gutmann" 35 pass method. They caution that the Gutmann method may take several weeks. But does anyone know how long the 7 pass US DoD 5220 will take? Thanks!
It'll take some time...primarily depending on the RPMs your drive is at (other factors as well). I'd expect about a day or so (dod3 takes us about 6-10 hours for that drive size where I work). Honestly unless you're protecting highly classified top secret material dod3 should be fine. The more passes you do the more sophisticated the equipment needs to be to recover it. At dod 7 you're talking about some expensive efforts.
posted by samsara at 8:26 PM on February 6, 2008
posted by samsara at 8:26 PM on February 6, 2008
This page shows a 10 pass KillDisk run at 30 minutes per gigabyte.
http://www.its.niu.edu/its/sh/downloads/wipedisk.shtml#etoc
I ran the single zero pass on a bunch of hard drives of varous sizes in about 10-15 minutes.
posted by so_ at 8:26 PM on February 6, 2008
http://www.its.niu.edu/its/sh/downloads/wipedisk.shtml#etoc
I ran the single zero pass on a bunch of hard drives of varous sizes in about 10-15 minutes.
posted by so_ at 8:26 PM on February 6, 2008
If possible, just run a single pass - 7-fold overwrite is extreme overkill. At even 1 or two passes, it will take forensic-level labs a very long time to get anything off of your drive.
One more data point - for an old 250 GB drive of mine that I wiped (7200 rpm), it took at least 12 hours for a single pass.
posted by chrisamiller at 8:32 PM on February 6, 2008
One more data point - for an old 250 GB drive of mine that I wiped (7200 rpm), it took at least 12 hours for a single pass.
posted by chrisamiller at 8:32 PM on February 6, 2008
it depends on how big the drive is and on the drive's maximum sustained write speed. drive size multiplied by sustained write speed multiplied by 7 is your lower bound. be aware that (according to wikipedia, at least)
posted by russm at 9:22 PM on February 6, 2008
As of the June 2007 edition of the DSS C&SM, overwriting is no longer acceptable for sanitization of magnetic media; only degaussing or physical destruction is acceptable.do you actually need to meet DoD 5220.22-M, or is this just a checkbox for a manager's procedural compliance?
posted by russm at 9:22 PM on February 6, 2008
Well, I am stating the obvious, but does this software have some kind of progress meter? Why not just let it run for n minutes and then do some arithmetic? If you're wiping the disk anyway, it's not like you're going to lose anything if you interrupt it and start over. I don't see much reason the progress would be non-linear.
And this is tangential to the question, but security only really makes sense in relation to the likelihood of a particular attack. I would seriously consider whether there would be any reasonable attack that would involve trying to read a wiped disk vs. other, easier ways of obtaining the same information.
posted by dixie flatline at 11:08 PM on February 6, 2008
And this is tangential to the question, but security only really makes sense in relation to the likelihood of a particular attack. I would seriously consider whether there would be any reasonable attack that would involve trying to read a wiped disk vs. other, easier ways of obtaining the same information.
posted by dixie flatline at 11:08 PM on February 6, 2008
I do computer forensics for a living. Anything more than one pass is overkill, unless you think that your data is worth enough for someone to spend millions of dollars over the course of a year to try to recover it - and still have the risk of failing. Do the minimum and don't worry about it. One pass is enough to defeat almost anyone. The government specs are paranoid to an extreme.
posted by procrastination at 4:51 AM on February 7, 2008 [1 favorite]
posted by procrastination at 4:51 AM on February 7, 2008 [1 favorite]
« Older i DO have a chip on my shoulder but I'd prefer it... | What do I do with a pack of 25 young teens in the... Newer »
This thread is closed to new comments.
posted by caddis at 8:24 PM on February 6, 2008