Announcement

Collapse
No announcement yet.

OpenZFS Is Still Battling A Data Corruption Issue

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

  • #51
    Originally posted by Developer12 View Post
    Disappointed to see Michael double-dipping on this news. Post an article about the bug, wait a few days, post another article with basically no change to the situation.

    It hasn't been fixed, it's still being root-caused, but he just can't seem to resist the traffic spike.
    Maybe I could accept that point if the head line of the first news would be obvious, but I like probably many read over it, because it starts "Openzfs released..." then the 6. world or so is bug and the 10. Data 11. loss.

    So as non user and not much interest in using it when I see a headline Openzfs released I don't care to much to read further and assume, ok they released something, and there is a list of new features.

    It's basically 2 news in 1 headline, Openzfs released and Openzfs has bugs including versions before that release.

    So I would not have noticed it at all if he would not had this 2nd news or much later indirect I maybe would have heard about it.

    Yes Michael might care about traffic to some degree, the question is did he something immoral what damage does that 1. again if he wanted to scandalize that the 1. news would be much more alarming and direct and clear starting with "Data destroying Bug..." or something.

    So what is the damage if somebody reads that now 2 times instead of 1 time? Now the undeserved myth that ZFS is 1000 times better than BTRFS and unproblematic vs the totally "unusable" btrfs is at least damaged if not gone, but with 1 clear news that would also be the case.

    And that this has already more comments than the 1. news seems to show that I am not the only one that did not register the news from this convoluted indirect report as side note of a bad bug. And of course the longer they need to fix it, the bigger of a problem it is the more fundamental more structural...

    EDIT:
    Also this is not exclusive to ZFS, if he with his testing or somehow else notices a bigger bug he very often has followup news to it.

    Comment


    • #52
      You guys jest but rewriting it in Rust would at least solve the license issue wouldn't it?

      Comment


      • #53
        Originally posted by BreachScrambler View Post
        You guys jest but rewriting it in Rust would at least solve the license issue wouldn't it?
        It would. Just like an independent reimplementation from scratch in any other language would.

        Comment


        • #54
          Originally posted by blackiwid View Post

          Maybe I could accept that point if the head line of the first news would be obvious, but I like probably many read over it, because it starts "Openzfs released..." then the 6. world or so is bug and the 10. Data 11. loss.

          So as non user and not much interest in using it when I see a headline Openzfs released I don't care to much to read further and assume, ok they released something, and there is a list of new features.
          Your entire point is moot because this exact same news was already covered a few days ago.
          Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

          Comment


          • #55
            Originally posted by Almindor View Post
            nobody understands
            except the experts who understand, and have fixed things.

            You're welcome.

            Comment


            • #56
              Originally posted by Mario Junior View Post
              This is a CVE now ​​​​​​
              https://github.com/openzfs/zfs/pull/...ent-1826780399 was a formality, for awareness. It generated some amusement. Please note, I wrote:

              I lean towards treating it … not requiring a FreeBSD security advisory (SA).​
              Some amusement, but I keep an open mind.
              Last edited by grahamperrin; 28 November 2023, 10:46 PM. Reason: open mind

              Comment


              • #57
                Originally posted by Developer12 View Post
                It hasn't been fixed, it's still being root-caused, but he just can't seem to resist the traffic spike.
                Please, don't be mean-minded.

                In my own words:

                openzfs/zfs issue 15526 is the most interesting computer-related issue that I have encountered in my lifetime. Possibly the most complex.
                Now:

                zfs-2.2.2 patchset by tonyhutter · Pull Request 15602 · openzfs/zfs
                Last edited by grahamperrin; 28 November 2023, 10:34 PM. Reason: PR 15602

                Comment


                • #58
                  Originally posted by ultimA View Post
                  It would. Just like an independent reimplementation from scratch in any other language would.
                  There's only two languages allowed in the kernel. Why bother with a rewrite if the goal is not to include it there.

                  Comment


                  • #59
                    Originally posted by grahamperrin

                    Sigh. Why do you have Developer in your username?

                    It's not the same.



                    Developer12 ​It seems to me that you didn't even bother to drive by.
                    Should michael write an article about every issue opened on their github about this bug? there have been over five of them and counting. How about for every comment? Every time a ZFS developer weighs in?

                    He's just double-dipping on the same ZFS bug twice in the span of a week because ZFS/BTRFS flamewars generate tons of engagement, and thus ad revenue.

                    Comment


                    • #60
                      I have a lot of data on ZFS both professionally and personally. I will keep using ZFS because there are no better alternatives if you need a feature set similar to what it provides.

                      If you don't need those features then consider not adding this complexity to your storage solution. It increases the risk of other types of issues. Professionally I stay at least a year behind the latest version, similar to what I do with other software unless there is a critical security issue.

                      It's a bit amusing to read that some people are shocked that the developers don't have a perfect overview and understanding of the details of the entire system. Developers who have worked on any project of similar size and complexity should know that over the years (and decades) some people will leave a project and others forget what they themselves wrote in the past. While project leads may have a pretty good overall understanding of how the system works it is impossible for any person to keep track of every little change added by everyone and their unintended combined side effects.

                      Comment

                      Working...
                      X