Not logged in
Log in now
Create an account
Subscribe to LWN
LWN.net Weekly Edition for May 23, 2013
An "enum" for Python 3
An unexpected perf feature
LWN.net Weekly Edition for May 16, 2013
A look at the PyPy 2.0 release
See, among other sources, http://web.archive.org/web/20090616233917/http://sansfore... and https://www.infosecisland.com/blogview/16130-The-Urban-Le... . I've also previously seen substantial prizes offered for the demonstrated ability to recover data from a drive overwritten once with zeroes, none of them ever awarded.
Securely deleting files from ext4 filesystems
Posted Oct 13, 2011 21:20 UTC (Thu) by sciurus (subscriber, #58832)
Posted Oct 14, 2011 0:38 UTC (Fri) by njs (guest, #40338)
Posted Oct 16, 2011 3:04 UTC (Sun) by giraffedata (subscriber, #1954)
I don't think I'd cite that paper. It is poorly written, from the disorganized presentation of thoughts to the poor grammar (I lost count of the sentences with no verb). That makes it hard to detect the theoretical basis for prior data not being recoverable.
Maybe that doesn't matter because the paper also reports empirical results, with numerical conclusions of the probability of recovering prior data that are insignificantly small. However, amazingly, the paper does not describe the method. I.e. it does not say how the investigators attempted to recover the data. (Maybe that is embedded in the ungrammatical and to me unintelligible sentence, "both drive skew and the bit was read.").
One thing the paper sort of hints at is the reason I believe the conclusion -- that you can't recover prior data from an overwritten disk -- is true. In the quest for greater density of current data, disk drive engineers have squeezed out whatever redundancy used to allow a drive to store both the current data and some of the prior data.
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds