Surviving a Linux Filesystem Failures

last updated in Categories Debian Linux, File system, Gentoo Linux, Howto, Linux, RedHat/Fedora Linux, Suse Linux, Ubuntu Linux

When you use term filesystem failure, you mean corrupted filesystem data structures (or objects such as inode, directories, superblock etc. This can be caused by any one of the following reason:


* Mistakes by Linux/UNIX Sys admin
* Buggy device driver or utilities (especially third party utilities)
* Power outage (very rarer on production system) due to UPS failure
* Kernel bugs (that is why you don’t run latest kernel on production Linux/UNIX system, most of time you need to use stable kernel release)

    Due to filesystem failure:

    • File system will refuse to mount
    • Entire system get hangs
    • Even if filesystem mount operation result into success, users may notice strange behavior when mounted such as system reboot, gibberish characters in directory listings etc

    So how the hell you are gonna Surviving a Filesystem Failures? Most of time fsck (front end to ext2/ext3 utility) can fix the problem, first simply run e2fsck – to check a Linux ext2/ext3 file system (assuming /home [/dev/sda3 partition] filesystem for demo purpose), first unmount /dev/sda3 then type following command :
    # e2fsck -f /dev/sda3

    • -f : Force checking even if the file system seems clean.

    Please note that If the superblock is not found, e2fsck will terminate with a fatal error. However Linux maintains multiple redundant copies of the superblock in every file system, so you can use -b {alternative-superblock} option to get rid of this problem. The location of the backup superblock is dependent on the filesystem’s blocksize:

    • For filesystems with 1k blocksizes, a backup superblock can be found at block 8193
    • For filesystems with 2k blocksizes, at block 16384
    • For 4k blocksizes, at block 32768.

    Tip you can also try any one of the following command(s) to determine alternative-superblock locations:
    # mke2fs -n /dev/sda3
    # dumpe2fs /dev/sda3|grep -i superblock
    To repair file system by alternative-superblock use command as follows:
    # e2fsck -f -b 8193 /dev/sda3

    However it is highly recommended that you make backup before you run fsck command on system, use dd command to create a backup (provided that you have spare space under /disk2)
    # dd if=/dev/sda2 of=/disk2/backup-sda2.img

    If you are using Sun Solaris UNIX, see howto: Restoring a Bad Superblock.

    Please note that things started to get complicated if hard disk participates in software RAID array. Take a look at Software-RAID HOWTO – Error Recovery. This article/tip is part of Understanding UNIX/Linux file system series, Continue reading rest of the Understanding Linux file system series (this is part III):

    • Part I – Understanding Linux superblock
    • Part II – Understanding Linux superblock
    • Part III – An example of Surviving a Linux Filesystem Failures
    • Part IV – Understanding filesystem Inodes
    • Part V – Understanding filesystem directories
    • Part VI – Understanding UNIX/Linux symbolic (soft) and hard links
    • Part VII – Why isn’t it possible to create hard links across file system boundaries?

    Posted by: Vivek Gite

    The author is the creator of nixCraft and a seasoned sysadmin, DevOps engineer, and a trainer for the Linux operating system/Unix shell scripting. Get the latest tutorials on SysAdmin, Linux/Unix and open source topics via RSS/XML feed or weekly email newsletter.


    33 comment

    1. Its easy to trash the old fs and
      create a new one , because data
      is never “hooked” to a particular
      fs !
      LiLo , LoadLin et al will still
      boot linux fine .
      Its easy cause the IDE HDD is
      much easier to figure , there
      is no need for partitions , nor
      any other “dependencies” .
      And drivers are much smaller .

      The reason we dont improve ,
      is someones paycheck is in it ,
      to keep it complicated .
      Its not unlike what Bill Gates
      did to USB …

    2. If I understood correctly there are typos within the following block:
      || The location of the backup superblock is dependent on the filesystem’s blocksize:
      || * For filesystems with 1k blocksizes, a backup superblock can be found at block 8193
      || * For filesystems with 2k blocksizes, at block 16384
      || * For 4k blocksizes, at block 32768.
      If first block is moved by one (8192 +1) then others must be moved by one also (16384 +1 and 32768 +1) , therefore backup superblock for 2k and 4k block sizes must be at 16385 and 32769.
      Is that right?

    3. can’t read superblock

      max@max-laptop:~$ sudo fsck -f /dev/sdd1
      fsck 1.41.4 (27-Jan-2009)
      dosfsck 3.0.1, 23 Nov 2008, FAT32, LFN
      There are differences between boot sector and its backup.
      Differences: (offset:original/backup)
      1) Copy original to backup
      2) Copy backup to original
      3) No action
      ? 3
      Both FATs appear to be corrupt. Giving up.

      Hummm… WTF… what can i do.

    4. “That works great IF your system will give you a terminal window. If you don’t get that far, you are SOL.”

      That’s what Live CDs are for… 😉

      Thanx for this page-I was able to completely recover a borked system. A great resource!

    5. hello!
      every command i run results in this message:
      e2fsck: attempt to read block form filesystem resulted in short read while trying to open /dev/sda1. Could this be a zero-length partition?

      i can determine alternative superblock locations but the command e2fsck -f -b 32768 /dev/sda1 + with other superblock locations doesn’t work.
      please help thanks !!

    6. Hi,

      I tried some data recovery programs and they all found the files on /dev/sdc3 BUT

      ANY attempt to “fsck -f -b -y /dev/sdc3” results in

      /: Attempt to read block from filesystem resulted in short read while reading block 525

      /: Attempt to read block from filesystem resulted in short read reading journal superblock

      fsck.ext2: Attempt to read block from filesystem resulted in short read while checking ext3 journal for /

      any Idea before I jump out of the window ?

      many thanks

    7. Several comments:
      * I was most interested in learning more about a superblock, and that part of the series was very helpful to me.
      * I haven’t started digging into it, but I’d like to learn more about how a corrupted filesystem is fixed, my specific question is, are there circumstances in which a filesystem is fixed by deleting all or part of a file? (I know this was done in early versions of Dos and Windows, but I haven’t done much with either in the last 12 years, maybe that is no longer the case.) But I want to find out if that is, or ever was, the case in Linux.
      * Your tutorials are very well written, but you have some typos / grammos. If you’re interested, I can send you (sooner or later) copies of the content with the errors I see fixed. (Just as one example, the title of this page has a problem–you’re mixing singular and plural stuff, the title should either be: “Surviving Linux Filesystem Failures” (all plural) or “Surviving a Linux Filesystem Failure” (all singular).

    8. No matter what I do, I get following msg.

      root@tdsrv002 [~]# e2fsck -f -b 32768 -y /dev/xvdj
      e2fsck 1.41.12 (17-May-2010)
      e2fsck: Bad magic number in super-block while trying to open /dev/xvdj

      The superblock could not be read or does not describe a correct ext2
      filesystem. If the device is valid and it really contains an ext2
      filesystem (and not swap or ufs or something else), then the superblock
      is corrupt, and you might try running e2fsck with an alternate superblock:
      e2fsck -b 8193

      I tried all block locations which I found here.

      root@tdsrv002 [~]# mke2fs -n /dev/xvdj
      mke2fs 1.41.12 (17-May-2010)
      Filesystem label=
      OS type: Linux
      Block size=4096 (log=2)
      Fragment size=4096 (log=2)
      Stride=0 blocks, Stripe width=0 blocks
      1966080 inodes, 7864320 blocks
      393216 blocks (5.00%) reserved for the super user
      First data block=0
      Maximum filesystem blocks=4294967296
      240 block groups
      32768 blocks per group, 32768 fragments per group
      8192 inodes per group
      Superblock backups stored on blocks:
      32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,

      Thanks for any hint to fix this issue.

      1. Hi,

        Did you find way to fix this issue. I’m facing same problem. Help me to recover data in HDD.


    9. So far, sounds useful. My issue is not the root FS, so any live CD isn’t needed. My issue at the moment is the USB hosted external HDD. By the time I see a reply, I will probably have made a decision that finalizes the matter, so this post is mostly about educating the ‘future reader’.

      While manipulating files in KDEs Dolphin (KDE 4.7.4), I suddenly lost access to a directory on that external HDD. The directory holds valuable files, naturally, and this IS my backup device. FWIW, I’ve seen KDE having many issues in my mode of operations, no matter whether on a backup device or on a day-to-day flash device. In this instance, I’m pretty sure I botched it myself.

      Dolphin file manager allows a bash prompt child window, and I tend to use this from time to time – I often need to change permissions, owners, etc. of files that I have collected over the years. An example is where I’m mixing files over several years storage, so I see problems that I can only guess are related to user group for my normal username being group ID 101 for Mandrake files, and 1000 for current Kubuntu systems. All I can tell is that everything ‘looks’ correct, from what Dolphin displays, and yet files don’t open, or pop up an error message when moved within Dolphin. I then look at permissions within Dolphin, and clearly see a problem that I can fix. In the latest error, I needed to issue chmod 755 for a directory of files. I’ve tried using the Dolphin GUI to do this in the past, and seen it refuse to act despite displaying no errors, so today, as learned from the prior experiences, I used the bash propmt at the child window, and issued the chmod command. The prompt returned almost immediately, bu I neglected to sync the changes. I then closed that Dolphin window, and I thnik that caused the chmod to get zombied. In any event, the files and the directory thet are in cannot be accessed.

      I have run FSCK -y -b xxxx /dev/sdc1 for EVERY superblock I found, no joy. Despite displaying many thousands of correction messages:
      Free inodes count wrong for group #371 (8192, counted=8180).
      Fix? yes
      Free inodes count wrong for group #372 (8192, counted=8190).
      Fix? yes
      Free inodes count wrong for group #373 (8192, counted=8186).
      Fix? yes
      Free inodes count wrong for group #374 (8192, counted=8190).
      Fix? yes
      Free inodes count wrong for group #375 (8192, counted=8186).
      Fix? yes
      Free inodes count wrong for group #376 (8192, counted=8150).
      Fix? yes

      After these positive sounding messages, I mounted the device, still could not get into the directory, unmounted the device, and repeated the fsck with a new superblock but cannot even ls -la the directory from a Konsole.

      Pretty sure I killed it when I didn’t look (in Dolphin) for the chmod command results on that directory. But, any suggestions on what could be done when fsck achieves no improvement?

    10. Is there any way that we can predict a file system corruption?
      And/or is there any tool that we can use?

      Open for any suggestions.
      Thank you so much!

    11. Thank you so much, i have same problem with the file system corruption & i try with command line # sudo e2fsck -f -b 8193 -y /dev/sda6 –> WORKING …
      that you so much is very very helpful …
      Finally I can get my data …

    Leave a Comment