Harddrive Failure and Data Recovery

  • [I don't know why, but every time I submit a Reply, I get an Error Occurred. But the Reply goes through anyway.]



    Boot into the BIOS and see if the all the disks show there. If not, find out why.


    You're right gderf. In BIOS, two SATA ports show "none". And with SATA 0-4, it's the SATA2 and SATA3 that are blank. I'll do some investigation and see if there's anything obvious, like a bad connector or something.


    One other thing I was wondering ... since three data disks were unionfs before, will that make any difference about seeing data? I have the Paragon Software on my PC that allows me to look at the drives in Windows File Explorer. Shouldn't the content of files be seen in there somewhere?

  • First thing I would look at are the power cables - do the disks spin up and get warm?


    Unionfs (mergerfs) just presents a pool of disks. It doesn't alter the data on them - each disk stands alone so you can look at them individually.

    --
    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.

    • Offizieller Beitrag

    I'm not sure which attributes trigger the red light, but it's not good.

    Red light indicates bad sectors, but don't know the attribute, it would suggest those drives are in pre fail, sooner rather than later.

  • So hosed :( .
    The two drives that don't show up will not spin up in my Paragon Software either. Each one does about 12 clicks and then spins down. Obviously some sort of sector problem. One of the other two drives that OMV recognized won't mount either, even though in BIOS it showed that one as a SMART drive.


    Looks like I'm going to have to find a wizard to get into these drives and see if I can recover my data.


    :(:(:(:(:(:(

    • Offizieller Beitrag

    Each one does about 12 clicks and then spins down.

    Dead, probably the actuator, and you'll find all sorts of suggestions from placing the drive in a zip lock and putting it in the freezer for a couple of hours :)


    Looks like I'm going to have to find a wizard

    I don't think Harry Potter does house calls ^^

    • Offizieller Beitrag

    Hey, I would still try the BTRFS driver for windows, and see if you can look at those drives. You could have a 12V power supply problem which, if the PS is marginal, may affect different drives differently.


    While it's possible, I find it hard to believe that two drives would fail at the same time, together.

  • It IS possible that they died at about the same time. These are old drives from a previous Windows Home Server box manufactured in 2011. Somehow or another, I need to get the data off of these things. Then I can rebuild my OMV with my new four 2TB drives.


    The other day, I was able to see one of the drives in my Windows 10 using the Paragon Software, but now I can't get it mounted even though the software can see the drive.


    I am currently Googling for drive recovery tools or services.


    Yeesh!

    • Offizieller Beitrag

    You know,, I think this is going to be a good (or bad depending on the perspective) learning experience for you. A $30 SBC and a big external hard drive, for backup, must to be looking good right now.


    It wouldn't hurt to try another method of reading a BTRFS disk, with a windows machine. BTRFS driver


    I can say this, and you probably won't want to hear it - if those disks were formatted EXT4, getting the viable data off of them would be much easier. EXT4 is known for fsck and it's supported by data recovery packages. BTRFS?, not so much.

  • Hello,


    If I may summarize:
    1) you have two failed drives. Snapraid saves you from one failed only
    2) It is not clear to me, if one of the drives that has failed was a parity drive


    My recommendation:
    1) You should be able to just put in the working drives one by one to OMV mount them and recover the files by copying (not moving anyting around!!) to another drive. No messing with Snapraid, no messing with Unionfs nothing.
    2) For the failed drives, I recommend trying to keep it simple. Ensure that you get them back to life. If you don't the data on them is gone. If you get back one of them, you have a good chance to recover all data (also the one of the other) thanks to snapraid.



    For (2):
    First check a different system in order to make sure that it is not faulty power supply, cables etc. You could also use an external drive enclosure.
    Once you can see the drive in the BIOS of your other system, you can continue. If not, you need a professional data recovery service if you value your data. If not, you can try the freezer or so...
    Once you have the drive in the BIOS get another drive of at least the same size and a linux system (any linux will do) and use ddrescue to copy the content of the original drive to the new drive.
    The syntax is:


    Code
    ddrescue SOURCE TARGET /logfile.log
    e.g. 
    ddrescue /dev/sdc /dev/sde /logfile.log

    You need to be absolutely sure about the source and the target. Otherwise you will copy the content of your empty drive over the content of your valuable data and it will be gone forever.


    After this step, you have recovered your broken data from the broken drive. You can do the same with the second broken drive.
    If data is damaged after this, you can let snapraid try to fix it.


    If snapraid reports too many errors, you can send your original drives to a data rescue company.
    If your data is valuable, I recommend you to directly send both drives to a data rescue company and not to mess anylonger. Ask them to recover make you an offer to recover each of them. One of them should be enough. Preferably the data, not the parity drive.


    Greetings,
    Hendrik

  • Thanks! You guys are great!

    You know,, I think this is going to be a good (or bad depending on the perspective) learning experience for you. A $30 SBC and a big external hard drive, for backup, must to be looking good right now.

    Definitely a learning experience. I have already procured an 8TB external backup drive to use when all of this is resolved and I have four 2TB drives in my box. I have determined that the drives have gone south, but am hopeful I can get data recovered by a service. The box and MOBO are good.


    It is not clear to me, if one of the drives that has failed was a parity drive

    Yep, one of the failed drives was the parity drive.


    I have tried Paragon Software which initially read one of the data drives, but now won't "mount" the drive although it can see it. The first time I tried using Paragon, it did mount the drive and I was able to see multiple files. Sadly, I should have just copied everything into an auxiliary drive I have and investigated from there - don't know what was happening to cause me to not do that.


    I have tried using the btrfs driver item, but it doesn't show in Device Manager and Harmstone says he has experienced the need to install multiple times in order to see it. But that hasn't worked for me. Also WinMD can't find a winmd.sys file so it won't load - using WinMD was advice from Harmstone as well.


    Well, although this is has been a long and winding road, I am not giving up. Somehow this will turn around. I'll keep at it.

  • Stop Messing around with Windows and Recovery Software.


    Use Linux to Recover the faules Data Drive.


    As the parity Drive is dead, Snapraid cannot Help unless we can recover the parity Drive. But that would be the Second try.


    If you are willing to follow my advice, I am willing to Help.

  • Go to my pre previous Post and digest it. Ask Questions until you have fully understood.

    What is the method to properly identify whether a drive is sdb or sdc or ???
    What I have in there now is a spinning original 1TB drive with the data (hopefully) and a second 2TB drive that is blank. Both are BTRFS.


    Edit:
    Got it. Went into the WebGUI and discovered which one was sda and sdb.

  • You should be able to pick out what you need by running:


    tree /dev/disk -a


    Or you can look at OMV's Storage | Disks page.

    --
    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.

  • Now I need to find out how to safely turn off mei-me. Googled it, and some say use rmmod while others say use modprobe. So far the better, clearer, instructions are for rmmod.


    What do you say?
    (the main reason I want to turn this off is because it keeps interfering with typing into the command line)


    rmmod mei-me vs. modprobe -r mei-me

  • rmmod


    Background:
    bbs.archlinux.org/viewtopic.php?id=168403


    Hah! That is exactly the page I had looked at!


    rmmod it is


    Thanks henfri.
    ********************
    So, I tried to rmmod mei-me after logging into the root of OMV and it gave me a "module mei_me is in use" errror. Then I tried ddrescue and it said "command not found".


    Do I assume I should break out at the Debian screen and not let it boot into OMV ?
    ************************


    OK. I think I figured out the Linux issue. I am just downloading Debian Linux onto a USB stick and then will boot from there.


    Hope that's the right approach.

  • Hello,


    sorry, but I did not see your edit, as I did not get any notification for thouse.


    Regarding the module removal:
    https://askubuntu.com/question…el-module-which-is-in-use
    (I found that quite easily: https://www.google.com/search?q=module+is+in+use+rmmod)



    Zitat

    Do I assume I should break out at the Debian screen and not let it boot into OMV ?


    That is a misunderstanding. OMV are packages installed on top of Debian.
    You can also not stop Windows from booting into Microsoft Office...


    OMV is no problem for you.



    OK. I think I figured out the Linux issue. I am just downloading Debian Linux onto a USB stick and then will boot from there.

    No, don't worry about OMV. The module that is bugging you is not from OMV.
    I assume that you will have the same problem with Debian. And if not, that will be because you will be putting an older or newer version of debian onto your stick than the one that OMV is using. And then it will be harder for us to help you.



    Then I tried ddrescue and it said "command not found"

    But it's not so hard to find out how to install a command in Debian.
    You must install the software. It is called a package in linux. So you need to find out how the package is called that provides ddrescue.
    Google ddrescue install debian leads you to
    https://packages.debian.org/search?keywords=ddrescue


    and that tells you that there is a viewer (which we do not need ) and a package gddrescue which I would try.



    In debian you install packages with


    Code
    apt-get install PACKAGENAME



    So


    Code
    apt-get install gddrescue


    is our friend here


    Then, ddrescue will work (tried here).


    Please do not fall into the (common) trap of re-installing stuff. Usually it is easy to fix problems. Reinstalling stuff may fix the one, but bring another problem.



    Greetings,
    Hendrik

Jetzt mitmachen!

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