Announcement

Collapse
No announcement yet.

EXT4 Data Corruption Bug Hits Stable Linux Kernels

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #11
    So stable kernels aren't so stable.
    What else is new?
    That's why I run Debian... Kernel 3.2. Totally dodged this bullet.

    Have fun with this one folks

    Comment


    • #12
      I managed to get a kernel panic even with squeeze (2.6.32), the bug affected every 2.6.x/3.x kernel version.
      This is to say I'm fine with stable releases, you can't simply expect zero bugs.
      ## VGA ##
      AMD: X1950XTX, HD3870, HD5870
      Intel: GMA45, HD3000 (Core i5 2500K)

      Comment


      • #13
        Originally posted by Sidicas View Post
        So stable kernels aren't so stable.
        What else is new?
        That's why I run Debian... Kernel 3.2. Totally dodged this bullet.

        Have fun with this one folks
        Recently a change was made in ext4 in 2.6.32.y (used as many distributions' long-term support kernels) which broke GlusterFS:

        http://joejulian.name/blog/glusterfs...ucture-change/

        So it can and does happen... not very often, but yes.

        Comment


        • #14
          DA FUQ is wrong with the stable kernels ?

          data corruption on ext4 (again ?!)


          luckily I didn't check and update for 3.5.7 kernel yet since this is my main production box *argh*


          the more this happens the more I consider moving entirely to ZFS ...

          Comment


          • #15
            Originally posted by zman0900 View Post
            Any word on which versions of 3.4 are bad? I'm on Fedora 16 running 3.4.11 right now and haven't noticed any problems since my last reboot two days ago. I just installed the upgraded kernel 3.6.2 but luckly haven't rebooted to it yet. Thanks for the heads up, guess I'll be removing that update now.
            Here is a list of affected versions:
            • 3.4.14
            • 3.4.15
            • 3.5.7
            • 3.6.2
            • 3.6.3

            I have filed a bug in Gentoo's bug tracker for this issue:

            https://bugs.gentoo.org/show_bug.cgi?id=439502

            Comment


            • #16
              Welcome to linux

              Where the phrase "regression testing" was deleted from the test plan due to a disk error

              Comment


              • #17
                Originally posted by frantaylor View Post
                Where the phrase "regression testing" was deleted from the test plan due to a disk error
                The Linux kernel developers let distributions worry about regression testing.

                Comment


                • #18
                  Originally posted by ryao View Post
                  The Linux kernel developers let distributions worry about regression testing.
                  Apparently Linus Torvalds is actually just a bot that collects together kernel patches, it doesn't bother to check or review any of them.

                  When we see all those nasty emails from Linus, blasting crappy code, those are just computer-generated random messages designed to create buzz. In fact they just check it all in without looking at it.

                  Nope, the kernel developers don't do regression testing, the hold back kernel releases because they are too hung over from their partying. Yeah, that's the ticket.

                  Comment


                  • #19
                    Sh!t happens. I thank Ted for looking into the problem and patching it fast! +1

                    Comment


                    • #20
                      Originally posted by frantaylor View Post
                      Apparently Linus Torvalds is actually just a bot that collects together kernel patches, it doesn't bother to check or review any of them.

                      When we see all those nasty emails from Linus, blasting crappy code, those are just computer-generated random messages designed to create buzz. In fact they just check it all in without looking at it.

                      Nope, the kernel developers don't do regression testing, the hold back kernel releases because they are too hung over from their partying. Yeah, that's the ticket.
                      Feel free to help. If you think you can read and understand in every detail hundred thousand lines of code you can safely replace Linus.

                      Software has bugs. It is simply impossible to dodge them all. Just think about the notorious random number generator in debian some stable release ago....

                      Just thank you the openness of linux, will hit only a very small fraction of linux users and most likely geeks and contributors

                      Comment

                      Working...
                      X