Harddrive Failure and Data Recovery

    • Offizieller Beitrag

    Still want to know if there is some way to copy and paste these things


    Try installing PuTTY on a Windows box. In an SSH terminal window to your server, the right click of the a mouse is the equivalent of "paste" or "Ctrl+V"

  • dmesg |head

    [ 0.000000] microcode: microcode updated early to revision 0x27, date = 2019-02-26
    [ 0.000000] Linux version 4.19.0-0.bpo.4-amd64 (debian-kernel@lists.debian.org) (gcc version 6.3.0 20170516 (Debian 6.3.0-18+deb9u1)) #1 SMP Debian 4.19.28-2~bpo9+
    1 (2019-03-27)
    [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.19.0-0.bpo.4-amd64 root=UUID=673ed9a3-acc8-4039-b74f-72c5ce7d80bd ro quiet
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers'
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers'
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers'
    [ 0.000000] x86/fpu: xstate_offset[2]: 576, xstate_sizes[2]: 256
    [ 0.000000] x86/fpu: Enabled xstate features 0x7, context size is 832 bytes, using 'standard' format.
    [ 0.000000] BIOS-provided physical RAM map:
    [ 0.000000] BIOS-e820: [mem 0x0000000000000000-0x00000000000917ff] usable



    dmesg > /dmesg.txt

    Nothing



    dmesg

  • dmesg |head

    [ 0.000000] microcode: microcode updated early to revision 0x27, date = 2019-02-26
    [ 0.000000] Linux version 4.19.0-0.bpo.4-amd64 (debian-kernel@lists.debian.org) (gcc version 6.3.0 20170516 (Debian 6.3.0-18+deb9u1)) #1 SMP Debian 4.19.28-2~bpo9+
    1 (2019-03-27)
    [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.19.0-0.bpo.4-amd64 root=UUID=673ed9a3-acc8-4039-b74f-72c5ce7d80bd ro quiet
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x001: 'x87 floating point registers'
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x002: 'SSE registers'
    [ 0.000000] x86/fpu: Supporting XSAVE feature 0x004: 'AVX registers'
    [ 0.000000] x86/fpu: xstate_offset[2]: 576, xstate_sizes[2]: 256
    [ 0.000000] x86/fpu: Enabled xstate features 0x7, context size is 832 bytes, using 'standard' format.
    [ 0.000000] BIOS-provided physical RAM map:
    [ 0.000000] BIOS-e820: [mem 0x0000000000000000-0x00000000000917ff] usable



    dmesg > /dmesg.txt

    Nothing



    dmesg

    attachment

    cat /var/log/syslog | grep btrfs

    Nothing for this

  • What is sda?


    And where ist the raid6 coming from?


  • What is sda?

    sda is the bad disk

    And where ist the raid6 coming from?

    No idea. The only thing I can think of is this item in my BIOS (the O's under SATA Operation are radio buttons and the only one selected is RAID On):


    SATA Operation
    O - Disabled
    O - ATA
    O - AHCI
    O - RAID On


    This option configures the operating mode of the integrated SATA hard drive controller.


    Disabled = The SATA controllers are hidden.
    ATA = SATA is configured for ATA mode.
    AHCI = SATA is configured for AHCI mode.
    RAID On = SATA is configured to support RAID mode.

  • No...


    Please show the output of

    Code
    btrfs filesystem show

    Anyway anyway that clearly seems like the disk has a Hardware Problem. We should not further Stress it.


    I would now take this Disk out of the system.
    And work with the copy.

  • Hello,


    I was now able to have a more indepth look at your log.
    Please provide the output of

    Code
    smartctl -a /dev/sda


    and the same for the other drive that also showed errors.


    Then


    Code
    btrfs filesystem show -d

    (sorry, I should have written this in the first place. This will show all btrfs filesystems. Otherwise, you have to specifiy one)
    Post that output.


    Before we restore the data from the broken drive:
    You had also other drives, that were not damaged. Have you saved/backed up that data to another safe place?


    To now restore your data:

    • The scrub I would not do on the damaged drive, but on the copy that you did.
    • Please shutdown, remove the broken drive.
    • Put in another drive on which you can store the data we are going to restore. Can be an USB drive.
    • Start again.
    • Then, make sure that you properly identify the drive that we did copy the broken drive to first, in your omv-GUI (it may not be /dev/sda anymore). Lets say it is /dev/sdY
    • Properly identify the restore target (e.g. the USB drive) format it with a filesystem you can read in windows and mount it. Remember the mount point (e.g. /media/restoredrive/)
    • Unmount the drive in the OMV Gui, in case it is mounted
    • run btrfs restore /dev/sdY /mnt/restoredrive | tee /restorelog.txt

    That will run a while. After it has finished you can check what was restored on /mnt/restoredrive or shutdown and check the content of the drive in windows.
    Post the restorelog.txt.


    If this does not bring the success, we have two other options (btrfs scrub and btrfsck). But this one for sure does not change any data, so it is the safest, if properly followed.


    Greetings,
    Hendrik

  • smartctl -a /dev/sda

    As attachment.


    I will reinstall that other drive that was less "friendly" than the one in the attachment. Then I will provide the btrfs filesystem show -d that you instructed. I did not have any drives that were good, so I'm not sure what you are referring to here.


    I will get to the rest of the instructions soon.


    Stay tuned ...

  • smartctl -a /dev/sdb


    This attachment is for the 'other' drive that I had set aside that mounts but can't be read.


    BTW, we did not do a copy from this drive.



    btrfs filesystem show -d

    btrfs filesystem show -d
    Label: none uuid: c81bf277-6ded-4e03-8bce-d4b25a690e27
    Total devices 1 FS bytes used 384.00KiB
    devid 1 size 1.82TiB used 2.02GiB path /dev/sda1


    In this case, the "other" drive was installed. Results are showing our "copied" drive from ddrescue.


    btrfs filesystem show -d

    And this is with the one drive that we copied from. Again, only showing our 2TB copy drive.


    btrfs filesystem show -d
    Label: none uuid: c81bf277-6ded-4e03-8bce-d4b25a690e27
    Total devices 1 FS bytes used 384.00KiB
    devid 1 size 1.82TiB used 2.02GiB path /dev/sda1


    So, now on to the next steps you outlined.

  • Alas, I cannot find where to format a drive from the WebGUI. :(
    ***********************************************************************************


    Never mind, figured it out. :S


    **************************************************
    I must be seeing cross-eyed at this point.


    Properly identify the restore target (e.g. the USB drive) format it with a filesystem you can read in windows and mount it. Remember the mount point (e.g. /media/restoredrive/)

    I have formatted the restore drive to btrfs and mounted it. Is the mountpoint then /dev/sdb, which is what is showing in disks?

  • Storage | Filesystems press Create button, select device......................

    --
    Google is your friend and Bob's your uncle!


    OMV AMD64 7.x on headless Chenbro NR12000 1U 1x 8m Quad Core E3-1220 3.1GHz 32GB ECC RAM.

  • To all,
    I fear I may have to backup and start over at one of the levels. My 2TB drive that I ddrescue'd to will not mount. When I attempt to mount it in the GUI, I'm watching the command line output which indicates there is a BTRFS error - bad tree block starts, failed to read chunk root, open_ctree failed. The 1TB drive I ddrescue'd to (to fool around with in Windows) won't show up in the FileSystems list, and isn't recognized in my Windows system, which probably indicates that I did not format it correctly.


    So, I'm going to reformat the 1TB drive and ddrescue to that again, in the hopes that I'll get it right this time. Not sure what to do with the 2TB ddrescue'd drive at this point.

  • Don't do stuff that I am not asking you to, please.


    You have not understood the process.


    Birds-eye view:


    1) we make copies of the failed drive
    2) these are copies. They are not expected to be any better. But we can mess around with them
    3) we try to rescue (=mess around with the copies)


    Everytime you ddrescue, you stress your "about to completely fail" drive another time. Don't do that.


    You were (possibly) almost there. Just had to keep following my instructions from my last post.


    Greetings,
    HEndrik

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!