[sf-lug] recovery after dd

[Brian Morris]

now that I think about it, wouldn't the directory be overwritten with
all the names of the files ? you could recover the data by low level
read and EOF markers,  but you would still have no tree even, that is
you would also lose the directory names...

then there is the issue of fragmentation, where is the data on that
located, is it at the end of the fragment or is it in the directory
(filesystem) file ?

Brian





On 9/6/10, Dennis J Harrison Jr <dennisharrison at gmail.com> wrote:
> On Mon, Sep 6, 2010 at 12:18 PM, Sameer Verma <sverma at sfsu.edu> wrote:
>> On Sun, Sep 5, 2010 at 7:40 PM, Brian Morris <cymraegish at gmail.com> wrote:
>>> On 9/5/10, Sameer Verma <sverma at sfsu.edu> wrote:
>>>> I have an old external USB drive (120GB, I think) that had a bunch of
>>>> files and was close to full. I ran dd on it and installed a 2.1GB
>>>> image.
>>>
>>> I don't know about anyone else, but it is unclear to me
>>>
>>> a) exactly what the original condition of the drive was
>>>
>>> b) exactly what you did (arguments to dd ? what did you do by
>>> "installed" you say ?)
>>>
>>> c) where did you get the idea of doing this ? didn't they say how to
>>> the other half of the method ?
>>>
>>> Without this information, it would be hard for anyone to give any
>>> guidance as to what to do as far as I know.
>>>
>>> Brian
>>>
>>>
>>>
>>
>> Sending this out to Brian and the list. He forgot to cc list.
>>
>> Hi Brian,
>> This is an experiment. I had a external USB drive lying around (close
>> to full), so I'm playing with it. I also had a 2.1GB image from an
>> earlier experiment, so I used dd if=image.img of=/dev/sdc1 bs=1M but
>> what I want to figure out is if the data past 2.1GB is recoverable.
>>
>> Hope this clarifies.
>>
>> cheers,
>> Sameer
>>
>
> For what it's worth - this sounds like a blast.  Perfect thing to
> spend a beer session with on this holiday.  I'm filling up a 16gb
> drive right now to try this out.  My guess is that the remaining data
> will be recoverable.  I've very little knowledge of the fundamental
> workings of the rom, ram, nand and it's read / wrtie processes.  If
> the data isn't recoverable - I'll be reminded of just how little I
> know :)
>
> All the 1s and 0s on the other side of the partition should still be
> the same.  Unless for some reason the mass storage controller itself
> does something special when a partition is created (doubt) ?
>
> Essentially/Electronically speaking, electrons pass through two or
> more gates - one of them being a control gate.  The control gate
> passes electrons to a floating gate which then pushes the electrons
> through some oxide(ish?) layer.  The negative electrons then act as
> resistance.  There is a sensor which measures this to determine if it
> should report 0 or 1.  This happens per block (as far as I know, newer
> tech might do some neat stuff here and there).
>
> So, another few minutes on this transfer, then I've got to find a
> smaller drive I can dd to it.
>
>> _______________________________________________
>> sf-lug mailing list
>> sf-lug at linuxmafia.com
>> http://linuxmafia.com/mailman/listinfo/sf-lug
>> Information about SF-LUG is at http://www.sf-lug.org/
>>
>
> _______________________________________________
> sf-lug mailing list
> sf-lug at linuxmafia.com
> http://linuxmafia.com/mailman/listinfo/sf-lug
> Information about SF-LUG is at http://www.sf-lug.org/
>