On 20/09/11 14:50, Lee Winter wrote: > On Mon, Sep 19, 2011 at 10:23 PM, Scott Ferguson > <prettyfly.producti...@gmail.com> wrote: > >> I use Dban and shred (stick them in an old machine and take as long as >> it takes) - then disable the drive (pin in the breather hole), pliers on >> the power connectors.
^^ I'm refering to non-business data drives. > > DBAN is definitely one of the better tools out there, but it has > weaknesses that have to be considered. For example, it believes the > drive ID and info. It uses that info to determine what needs to be > done (e.g., number of sectors to be written). If the drive is working > and being replaced to increase capacity, that it not a problem. But a > drive being replaced due to unreliability or with intermittent errors > can deceive DBAN which will happily scrub only the number of sectors > reported by the corrupted firmware. > > So when you run it, particularly when doing batches of drives, you > have to verify that the ID and drive info matches the specs on the > drive. Noted.... I suspect that what the process that runs before DBan does. That is prior to "supervised destruction". DBan (or degaussing) is part of the "decommissioning process" Some sites use a "bang-box" instead of a wipe - the same logging and smelt procedures after though. Shred is used to delete files while a hard drive is functioning. For personal computers simply barbecuing the drive should render it unreadable except to agencies unlikely to have the time or motivation to recover data from it - though the pcbs should be removed first. For the truly paranoid - thermite (see youtube for dumb examples) or just use Truecrypt right from the start. If you have money to burn (for a new microwave), a pre-nuptial agreement *and have removed the pcbs* - a large microwave oven *will* wipe the data. (tested). Better to put into context though. In all instances strong encryption should be used. It's only after the fact you can safely determine what *was* necessary. If it contains government or business information - wipe the drive and securely dispose of it. If it only contains personal - wipe the drive and think hard before selling it or disposing it. Personally I wouldn't sell or give away a drive that contained personal information - and I'd recommend disabling it before disposable. Shredding or melting is probably overkill - but tossing on the barbeque/bonfire is quite reasonable - just remove the PCBs first. > > Lee Winter > Nashua, New Hampshire > United States of America (NDY) > I'm mostly addressing the subject - not you Lee. Some people seem to have misinterpreted Gutmman:- [quote] "In a followup to his paper, Gutmann said that it is unnecessary to run those passes because you cannot be reasonably certain about how a modern hard disk stores data on the platter. If the encoding is unknown, then writing random patterns is your best strategy. In particular, Gutmann says that "in the time since this paper was published, some people have treated the 35-pass overwrite technique described in it more as a kind of voodoo incantation to banish evil spirits than the result of a technical analysis of drive encoding techniques. As a result, they advocate applying the voodoo to PRML and EPRML drives even though it will have no more effect than a simple scrubbing with random data... For any modern PRML/EPRML drive, a few passes of random scrubbing is *the best you can do*". [/quote] Emphasis is mine. Sadly, people seem to believe that "a few random write passes is secure" - whereas our Kiwi researcher is just pointing out the limitation of overwriting data. The newer drives have tighter tolerances - additional writes simply repeat patterns and serve no purpose. NOT a couple of random write passes and the data can't be recovered. The ever decreasing footprint (magnetic field) of subsequent writes means that partial recovery of original data is always possible (not necessarily easy) - therefore, reconstruction is theoretically possible (but not certain). Extracting information from that "reconstructed" data is a lot harder - but in many cases only small amounts of information needs to be recovered to cause concern. Hopefully this thread has now run it's course... but I may be letting optimism triumph over experience (the world is full of fools who equate absence of evidence with proof of a negative) ;-p Cheers -- "Always question authority, and demand the truth." — Bill Hicks -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/4e784440.4000...@gmail.com
