Jump to content
Piriform Community Forums

Augeas

Moderators
  • Content count

    4,078
  • Joined

  • Last visited

Posts posted by Augeas


  1. Did you wipe free space using Drive Wiper? Drive Wiper overwrites the deleted records in the MFT so Recuva should return a long list of null, or ZZZ.ZZZ file names. All these null files should be a few hundred bytes in size, and be contained in the MFT. There shold be no clusters allocated to them.

    Are you still seeing the thumbnails? If you are, these should be found with a deep scan, and will have a numerical file name assigned instead of the actual filer name. These will not be overwritten by a live file, and can be securely deleted (which is just another overwrite).

    If you are seeing the thumbnails, go to the Info panel as described above and post a screenshot of what is there.


  2. Rob, if you look at any of the forums you will see that 'no one replies to anything' is nonsense, and only aggravates those who do reply. I understand that creating a disk image is a paid-for option, so you are entitled to direct support from Piriform. I have the free version of Recuva so I can't reproduce, or experience, your problem.

    You say that the screen goes blank, but fails the space calculation. Does it do both?

    If you're trying to recover a file or files, I would suggest that you don't run the disk image creation whilst there are problems with it.


  3. If you open an existing ExCel spreadsheet a temporary file is created called ~$Filename.xlsx. This is 165 bytes in size. If you open an ExCel file called Invoice #150.xlsx a file of 165 bytes is created called ~$Invoice #150.xlsx. Forget what I said about the recycler, this is (I believe) what you have recovered and are trying to open. I can create such a file and open it easily, so I don't know why you are having any difficulty. However, it isn't the file you want. The empty ExCel file I created to test this is 7.63 kb in size. No valid Excel file is 165 bytes in size.

    Forget this file, it will get you nowhere. It appears to be left behind after a crash of either ExCel or your pc at some point. You need to search for another ExCel file and. depending on how many you find, recover them to a folder on another drive and then look through them to see if you have found your invoice,. Don't bother with files of 165 bytes.

     


  4. I presume you can't rename it?

    The file is 165 bytes in size. I very much doubt that there's anything of use in there. Files sent to the recycler have their names changed to $Innnnnn.ext and $Rnnnnnn.ext. The $I file is an index and was (up to Win 10) 145 bytes in size. Perhaps this is what you have found. The $R component is the actual file data. That's the one you want.


  5. The larger disks get, the more difficult it is for both recovery software and the poor person handling the recovery to manage such large amounts of data.

    Whilst a pause or save facility might seem a good idea it brings the problems that if the pc is being used then the underlying data is being changed, it would taks some time to reload or locate previously found data, and worse of all Recuva would have to write a fair amount of data to the disk, which is something that it tries very hard not to do, as this can destroy what's trying to be recovered.

    Files can't be reconstructed or recovered in place. The complexities of modifying MFT records, MFT bit map, cluster bitmap, folder entries, and clusters (even assuming that they were still available for resurrection, would be horrendous. Add MFT extension records, index records etc and it's even worse. NTFS does not allow anyone to touch system metafiles either. Recovery to another device is just, well, safer by far.


  6. It probably has no, or very little, effect on the lifespan of the device. A retrim (which is what an SSD optimise is) will issue the same TRIM command to pages already trimmed, as NTFS - which issues the TRIM and retrim commands - has no knowledge of what pages have already been trimmed.

    I would imagine that the SSD controller would recognise that a page has already been trimmed and ignore the retrim command for that page. A cluster which has had a successful TRIM executed against it does not have a physical page allocated to it so it's difficult to see how it could be trimmed a second time.

    I've used the words probably and imagine as nobody here knows how the propietary software in SSD controllers works at this level of detail.


  7. It isn't easy to describe a problem and it certainly isn't easy to diagnose one from a distance. It would help if you could answer the questions asked:

    Is the file system FAT32 or NTFS?

    In Recuva Advanced mode select one of the thumbnailed files and in the info panel on the right:

    How many clusters are allocated to the file?

    How many clusters are overwritten?

    What is the name of the overwriting file?

    P.S. Don't pick the 2mb file, pick something around 20k.

    P.P.S. Confirm that the disk isn't an SSD, and isn't a shadow copy etc.


  8. Yes, a live file is one that hasn't been deleted. The thumbnails belong to the live file that has overwritten the deleted clusters. They do not belong to the deleted file (if the deleted file has been completely overwritten, that is). If you look at the comment alongside the file you will see what is overwriting the deleted file, and in Advanced mode you can see whether some or all clusters have been overwritten.

    You cannot get rid of anything that belongs to the overwriting file. It is live data.


  9. It means what it says, that the clusters of the deleted file have been overwritten by a live file. What you are seeing are thumbnails (or whatever) from the live files. You can't overwrite these clusters, they contain live data.

    I suspect that this is FAT32?

    If it's NTFS you could run Drive Wiper. This will wipe the MFT and you will no longer get these messages.

    By the way these messages are returned from the normal scan that a deep scan runs automatically as part of its process. Files found by the deep scan component (with a [001234].ext file name) will not, and can not, be overwritten by a live file.


  10. Not really. The file name will be in a folder somewhere (FAT or NTFS) and it looks as if it's been overwritten or corrupted somehow. The folder path can't be constructed (see the ? in the path) so what's happened to the folder I've no idea.

    As the file is in one extent you could use a freebie hex editor like HxDen and copy all of the clusters in one go, but that is quite a task. If it's FAT you could use HxDen to rename the file in the directory to something readable and then retry Recuva, if you could find the directory that is. You won't be able to do this with NTFS as the MFT is protected. Or you could - and this is what I would do first as it's easy and non invasive - run a deep scan. I don't think that files that are in folders are recognised by a deep scan but it's worth a try. A deep scan doesn't recognise file names.


  11. Mta, on a HDD you can use whatever character you wish, it will make no difference. As user data is scrambled and coded by the disk controller before being written to the disk there is no way to determine what sequence of ones and zeroes was used on writing nor what it will be on overwriting. So you can overwrite an unknown sequence with another unknown sequence, but you can never actually overwrite a one with a zero, or vice versa, as the scrambling removes the knowledge of whether a one or a zero has been, and is being, written. Overwriting a one with a zero is a logical construct far abstracted from the storage device.

    On an SSD you can't overwrite anything, you can only write new pages, so the proposition that you can overwrite with zeroes or ones or whatever, no matter how many times, is merely fanciful.

    To the SSD controller a page is either valid (i.e mapped to a lba and visible to the O/S) or invalid (i.e. not mapped to a lba and not visible to the O/S). This does not represent what the O/S considers to be live or deleted files. File life or death (and fragmentation as well, but that's another long bedtime story) to the O/S is specified in the MFT, or FAT tables. The SSD knows nothing of these.

    Deleted (TRIMed) SSD pages will be inaccessible to any operating system. Garbage Collection will empty invalid pages by setting all cells to 'ones', and those pages will remain inaccessible until the SSD controller receives a write request and decides to use them.

    DDR4 ram is volatile, but there are some hybrid cards with ram and non-volatile flash in the pipeline, perhaps this is what is meant. This is a long way from the O/P's suggestion.

     

×