0

a friend of mine currently has a problem with his external Harddrive. Unfortunately it's the only drive he's got a lot of old family photos on, so it's quite important for him.

Behavior and my steps til now were:

  • Plugging drive into Windows PC didn't show it in explorer
  • Windows Diskmanagement tells that it doesn't have valid MBR OR GPT (like in this picture)
  • Ubuntu 18.04 doesn't mount the drive
  • GNOME disk Utility shows the drive with 2,2TB even though it should only have 250GB (Model: ST9250315AS)
  • GNOME disk Utility shows "Disk is OK, one bad sector"
  • I ran ddrescue -S. This read through 2,2TB (?), but only had read errors. No data at all.

What further things could I try to save the disk? Do you have any idea why it's showing 2,2TB for the drive? Could it help to use a SATA-adapter to connect to the drive directly? I couldn't try that yet as I don't have one here currently.

EDIT: Seems like ddrescue wasn't finished yet. It went through the 2,2TB with around 167 MB/s error rate. Now it's starting over again with about 1200 kB/s error rate, counting up the bad sectors. Does it even make sense to wait for it to finish? Is there any possibility to point it to the correct 250GB first?

Output until now is:

    # Mapfile. Created by GNU ddrescue version 1.23
# Command line: ddrescue -S /dev/sdb /xxxxx
# Start time:   2022-01-20 18:47:42
# Current time: 2022-01-20 22:40:48
# Scraping failed blocks... (forwards)
# current_pos  current_status  current_pass
0x0FC0F600     /               1
#      pos        size  status
0x00000000  0x0FC0F800  -
0x0FC0F800  0x1FFF03F0600  /
0x1FFFFFFFE00  0x00000200  -
Improve this question
7
  • 2
    "What further things could I try to save the disk?" - Send it to a data recovery company, and they will diagnose the fault in the actual hardware, recovery of the data is not guaranteed. Your friend attempting to do anything will simply lower the changes of data recovery. In fact the actions already performed might have made it impossible to do data recovery on the drive.
    – Ramhound
    Commented Jan 20, 2022 at 21:56
  • Thanks for your comment. May I ask how the actions already performed might have made it impossible? Until now it was only read operations executed on the drive, or do I have a misconception here?
    – Guest
    Commented Jan 20, 2022 at 22:06
  • 1
    @Ramhound advice is VERY sound; however, not everyone is willing or has the means for this approach. If your other choice is "throw it away".. read on. First of all, this has nothing to do with the MBR. I myself would use dd to copy the drive to another drive before touching it again (AT ALL). Then with the NEW COPIED DRIVE, I would use TestDisk to recover what it can find. Keep in mind, dd has taken me WEEKS if there are many bad sectors so will need a dedicated machine
    – Señor CMasMas
    Commented Jan 20, 2022 at 22:11
  • 1
    Regarding your last comment, read operations still drag the needle over and try to read in bad sectors. In my experience, the more you do this, the worse the drive might become. I have 0% science or proof of this. I have LOTS of experience with this.
    – Señor CMasMas
    Commented Jan 20, 2022 at 22:13
  • Hi, thank you also for the answer! Doing a copy with ddrescue (is it the same as dd?) is what we wanted to try today, but as you said it's takind ages. For me that would be acceptable, but my fear is that it's because it's reading 2,2TB but it should only see 256GB. So the question is if I could limit it to only read the relevant 256GB (which would only need 10% of the time I assume)
    – Guest
    Commented Jan 20, 2022 at 22:15

1 Answer 1

Reset to default
0

What further things could I try to save the disk? Do you have any idea why it's showing 2,2TB for the drive? Could it help to use a SATA-adapter to connect to the drive directly? I couldn't try that yet as I don't have one here currently.

Assuming that you don't have an interface problem the best you can do is to stop using your drive immediately and contact a professional recovery lab.

There are two reasons for this recommendation: The incorrect size information of the disk (doublecheck using hdparm -i or smartmontools -a to see if that is hardware or software-based) and the big erroneous area that covers roughly half of the disk.

As for your question in your comment

ddrescue (is it the same as dd?)

No it is not. ddrescue has been programmed with the idea in mind to maximize the extraction of information in a short time despite being slowed down by read errors. You don't want continue reading sector by sector in an dammaged area when you could effortless recover data in another area of the disk. The different read behaviour of ddrescue compared to dd reduces further wear to a defective drive.

As for SenorCMasMas comment on the COUNT parameter in dd please be aware that disk drives do not return to the operating immediately when diagnosing a read error. A disk retries too much on its own. That is something that should be commanded explicitly by the operating system. Whenever it comes to use the COUNT parameter usely it wise as the drive firmware is already repeating enough with the exception for drives that are configured to work in RAID arrays.

Improve this answer

You must log in to answer this question.

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