0

My HDD is failing, the backup is made, only had 1 important file of 200 GB, I have currently 399 uncorrectable sector count and 399 pending sector count, plus 4874 reallocated sectors count. My hdd is 1000 GB.

I would like to know if I have lost something in my 200 GB file, including very small txt file of a few bytes.

Is there a way to know ? If not, how to calculate the odds of data lost in this file with the smart data ?

Thank you.

8
  • You would have to compare it to the original file in order to tell if the file was fragmented on any of those bad sectors.
    – Ramhound
    Aug 4, 2016 at 6:54
  • The problem is I don't have the original file size.
    – Chatai
    Aug 4, 2016 at 7:02
  • If you don't have an copy of the file, you know isn't corrupt, to compare the file then there is no way to know the copy you have is corrupt or not.
    – Ramhound
    Aug 4, 2016 at 7:08
  • If something is corrupt, even a very small text file, can the file disappear completely or does it always remain in its place, but the size become 0 byte or impossible to open ?
    – Chatai
    Aug 4, 2016 at 7:12
  • Depends what fragment of the file is corrupt.
    – Ramhound
    Aug 4, 2016 at 7:22

2 Answers 2

1

How did you back up your drive?

As Ramhoud explained if you don't have your original file and the data on that 200GB file is readable and executable there's no way of knowing if anything got corrupted or not. Usually, corrupted data can't be opened and would produce an error.

Is the old drive still in operation? Can you read the original data from it? Maybe you can see which sectors is the data taking and compare them with the problematic sectors on the drive and see if there's a possibility of some of the data being there.

2
  • I copy pasted my important files. I have my original file that might have lost data. The old drive is still powered on, I have access to the file.
    – Chatai
    Aug 4, 2016 at 13:19
  • In this case can you compare the two and see if anything is missing? Also check what paradoxon wrote!
    – Captain_WD
    Aug 5, 2016 at 12:13
0

First of S.M.A.R.T data is a bit of a mixed bag. It tends to be not outstandingly accurate / reliable. It is just reporting what the disk says. It is not asking any questions. It is useful as a warning sign.

Bad blocks doesn't always equal to lost data. Disks don't just write data and pray that never ever something goes wrong. The data is protected by forward error correcting codes and in many cases can be recovered from a bad block and moved. You can first try boot a linux distro and and us bad blocks command or use the tool packaged with windows for drive analysis and repair. Th badblocks command would be:

badblocks -vs /dev/sd#

Where # is the drive number. I would recommend booting a linux and us bad blocks since it is usually better to do this while the drive is not in use.

These tools are fine for a first try. If they fail there are more sophisticated solutions. My trusted one for years has been SpinRite. It is not open or free but it is very affordable and has been around for years. I would not use it to "resurrect" disks for prolonged us as people tend to do, but it is very useful for data recovery and the price is right. If your data is precious to you it might be worth spending the 90 bucks go give it a try.

GRC SpineRite

And for the future ( I am sorry but I have to stress this ), do make regular backups. It is not paranoid to do so. When done properly it is very cost effective counter measure to many problems. Disk failure, theft, fire, human error and even many types of virus infections. Take a regular full system backup (once you made major system changes or quarterly, you only need the most recent one) and make daily (if you are brave weekly) backups. If stuff hits the fan you can be back to normal within 30 min.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .