Announcement

Collapse
No announcement yet.

Arch Linux Powered CachyOS Now Defaults To Btrfs Rather Than XFS

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

  • #21
    No, please. Btrfs sucks. This is a step back and I'm totally against it. XFS is a lot more robust.

    Comment


    • #22
      Wasn't cachyos just cpu-optimized packages?

      It is n.g. if they start doing unnecessary extra changes. It does then become yet another derivative distro.

      Comment


      • #23
        Originally posted by timofonic View Post
        No, please. Btrfs sucks. This is a step back and I'm totally against it. XFS is a lot more robust.
        You can still pick XFS in the installer easily, without manually partitioning. It's just a drop-down that defaults to btrfs now instead of xfs. xfs is still there.

        Comment


        • #24
          Originally posted by timofonic View Post
          No, please. Btrfs sucks. This is a step back and I'm totally against it. XFS is a lot more robust.
          Yeah... gotta say LVM + XFS is pretty darn rock solid! You can achieve all BTRFS functionality in a much more modular method. That said... its a lot more modular and you do have to "lego build" it up where as BTRFS and ZFS have a much more unified CLI. Also not sure if LVM can do anything like ZFS RAIDZ... but also have not had a use case for that yet.

          Comment


          • #25
            Originally posted by intelfx View Post

            Correction: if you are prepared to throw entire racks of disks into your NAS, then ZFS is the obvious choice

            Btrfs is far more flexible than ZFS when it comes to changing the topology of your pool. If you have, like me, started with just a single HDD and later expanded that storage to 2, then 3, then 4 disks — then I just do not see how you would have used ZFS without recreating the pool at each point.

            In other words, ZFS might be infinitely more advanced (and don't get me wrong, it is), but all that is worth nothing if you can't use it within the physical (and, thus, financial) constraints you possess.
            Erasure coding (EC) plus Byzantine Paxos fault tolerance (BPFT) is even a lot more powerful and flexible, maybe even more robust and resistant to errors. The topology of most filesystems is terribly outdated.

            EC + BPFT has historically existed over forty years only in very big iron servers with very high end hardware and many times with redundancy too. Despite of that advantage, it proved to be a lot better than the rest over lots of years, because that paranoia is good for reliability.​

            I believe it'll be possible in the near future to have reliability, robustness, adaptability, flexibility and ease to use on the same filesystem.

            Comment


            • #26
              Originally posted by muncrief View Post
              I'm curious as to why the default filesystem was changed from ZFS to Btrfs. From what I've read Btrfs is much more prone to errors and far less mature than ZFS. Am I incorrect?
              Btrfs has always had broken RAID5/6 support, which makes ZFS a better choice for the typical home NAS use case. Other than that, incl for RAID10, btrfs is rock solid.

              Comment


              • #27
                Originally posted by intelfx View Post

                Correction: if you are prepared to throw entire racks of disks into your NAS, then ZFS is the obvious choice

                Btrfs is far more flexible than ZFS when it comes to changing the topology of your pool. If you have, like me, started with just a single HDD and later expanded that storage to 2, then 3, then 4 disks — then I just do not see how you would have used ZFS without recreating the pool at each point.


                In other words, ZFS might be infinitely more advanced (and don't get me wrong, it is), but all that is worth nothing if you can't use it within the physical (and, thus, financial) constraints you possess.
                People keep posting this and it's somewhat misleading. You don't need "entire racks of disks". You can start with 1 disk then add a 2nd later as a mirror. Then you just add more mirrors 2 disks at a time in a RAID 10. I don't see 2 disks at a time as an insurmountable financial burden when you need to expand your storage.

                Comment


                • #28
                  Originally posted by jacob View Post

                  Btrfs has always had broken RAID5/6 support, which makes ZFS a better choice for the typical home NAS use case. Other than that, incl for RAID10, btrfs is rock solid.
                  Btrfs RAID 10 is basically guaranteed to lose data when any 2nd disk fails, even if it's in a different mirror than the first failure.

                  Comment


                  • #29
                    The official T2 support sounds great. I wish the needed patches were upstreamed by now. I have a final Intel gen MacBook Pro sitting on a shelf and it's a PITA to deal with Linux on it. I think it still has EndeavourOS on it, but those patches aren't in their regular kernel, someone else builds a new ISO with them occasionally. There's an alternative Fedora version also but you need to update it in a special way to avoid pulling in the regular unpatched Fedora kernel and breaking things. This sounds more "just use it like normal".

                    Comment


                    • #30
                      Originally posted by pWe00Iri3e7Z9lHOX2Qx View Post
                      The official T2 support sounds great. I wish the needed patches were upstreamed by now. I have a final Intel gen MacBook Pro sitting on a shelf and it's a PITA to deal with Linux on it. I think it still has EndeavourOS on it, but those patches aren't in their regular kernel, someone else builds a new ISO with them occasionally. There's an alternative Fedora version also but you need to update it in a special way to avoid pulling in the regular unpatched Fedora kernel and breaking things. This sounds more "just use it like normal".
                      Actually, for working Wi-Fi there is still required from the user to either fetch the firmware from another mirror (mentioned in wiki) or extract it from the IPSW, see T2 Wiki for this.
                      Sadly, the firmware we can not distribute and we wont fight with apple.

                      But the Kernel Patches are included in the default kernel and we are in contact with the developers of these patches, so that we can ensure compatibility with the new major kernel versions.
                      The configuration is automatically done via our hardware detection.

                      Comment

                      Working...
                      X