Thanks for the congrats and sharing my thrill!

I have one correction to share in step #9:

Correction:
10% of 450MB = 45MB = 45000 (not 450000)
The 1 MB recovery file for this point has the following parameters:

$ dd if/dev/hda3 ibs=1024 skip=115000 count=1k >> /windowsD/recover115000

Later guys,

Peter

> 9) I began recovering 1 MB size files to edit with Midnight Commander.
> How I did this:
> Example: if an email was at the 10% mark, this represents 10% of 450
> MB= 45MB = 450000 which I add to where I started recording (in this
> case 70000). This give me its location from the start of /dev/hda3.
> Therefore when copying the 1 MB image of the 10% email, I change the
> parameters and filename so it looks like this:
> $ dd if/dev/hda3 ibs=1024 skip=520000 count=1k >> /windowsD/recover52000

On Sat, 2 Aug 2003, Peter E. Williams wrote:

> 
> 
> 
> Success!!!
> 
> I thought everyone would be interested how I recovered original emails
> from formatted space.
> 
> This is how I did it:
> 
> 1) While doing a full install of SuSE 8.2 on a SuSE 8.1 system, I juggled
> a few partitions and became intimately aware of where my old and new data
> was located.
> 2) I decided that my data could be on one of two locations, in the free
> space of my main directory / (/dev/hda1) and my new home directory /home
> (/dev/hda3)
> 3) I made note of the total size of the free space of the home directory
> (900 MB). The total partition size is 970 MB; therefore 70 MB went to all
> the files composing the home directory, minus my precious emails)
> 3) From root console, I segregated (preventing any copying to it) 
> /dev/hda3 by unmounting it ($ umount /dev/hda3)
> 4) I created two 450 MB image files of the free space of /home, one for
> the start and the other for the end of the free space
> 5) I created a blank recover file on my backup hard drive using an editor.
> Starting out, I called the file recover70
> 6) I copied each 450 MB image file to my backup hard drive (/windowsD) and
> called the file /windowsD/recover70
> This is how I did it from console:
> $ dd if/dev/hda3 ibs=1024 skip=70000 count=450k >> /windowsD/recover70
> explanation of this: 1024 byte of info is read at a time
>                      Information is only saved after 70 MB of data is read
>                      Information is stopped saving after 450 MB is read
> 7) I used Midnight Commander (mc) at console to search this huge file for
> keywords, specifically, the email address of the lost emails
> 8) Once the email address was located, I kept a handwritten record of its
> location by noting the top right corner % number, indicating how far the
> email was in the 450 MB file. I would make a note of the % each time I
> searched and found something. Before I knew it, I had a list of where the
> majority of the emails were located in the free space.
> 9) I began recovering 1 MB size files to edit with Midnight Commander.
> How I did this:
> Example: if an email was at the 10% mark, this represents 10% of 450
> MB= 45MB = 450000 which I add to where I started recording (in this
> case 70000). This give me its location from the start of /dev/hda3.
> Therefore when copying the 1 MB image of the 10% email, I change the
> parameters and filename so it looks like this:
> $ dd if/dev/hda3 ibs=1024 skip=520000 count=1k >> /windowsD/recover52000
> 
> 
> Believe me, it's painstaking, but I don't mind. The emails are valuable.
> 
> 
> Whew!!!!
> Very happy indeed!
> Peter :))))
> 
> 

Reply via email to