Announcement

Collapse
No announcement yet.

ZFS On Linux 0.8-RC1 Delivers Native Encryption, Direct I/O & More

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

  • ZFS On Linux 0.8-RC1 Delivers Native Encryption, Direct I/O & More

    Phoronix: ZFS On Linux 0.8-RC1 Delivers Native Encryption, Direct I/O & More

    Following the release of ZFS On Linux 0.7.10, the developers went ahead and released the big v0.8 release candidate. ZFS On Linux 0.8 is bringing a lot of new functionality...

    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

  • #2
    so exciting! Great news! I hope someday it will be without DKMS! I don't know why but I do not like it, but prefer over fuse BUT fuse got some updates recently, so might be worth trying!

    Comment


    • #3
      What about other things, like ZFS ACL's?

      Comment


      • #4
        Originally posted by BiG_NoBoDy View Post
        so exciting! Great news! I hope someday it will be without DKMS! I don't know why but I do not like it, but prefer over fuse BUT fuse got some updates recently, so might be worth trying!
        Not gonna happen, Fuse will never be as performant since most of it resides on userspace and is in no way related to dkms, in fact they can even move to Fuse and still require dkms because the code is out of tree.

        I think what you wish is for Zol to be integrated on the main tree but i dont think that could happen license wise until Oracle stop being asshats

        Comment


        • #5
          Originally posted by aht0 View Post
          What about other things, like ZFS ACL's?
          What you mean by ACL, as far as i'm aware they are working

          zroot aclinherit passthrough local
          zroot xattr sa local
          zroot acltype posixacl local

          or you mean another type of ACL that is not supported yet?

          Comment


          • #6
            Originally posted by BiG_NoBoDy View Post
            so exciting! Great news! I hope someday it will be without DKMS! I don't know why but I do not like it, but prefer over fuse BUT fuse got some updates recently, so might be worth trying!
            what's wrong with dkms?

            Comment


            • #7
              Has cache that doesn't clear after reboot been implemented or in the works? It's been on Solaris for years.

              Comment


              • #8
                Originally posted by jrch2k8 View Post

                Not gonna happen, Fuse will never be as performant since most of it resides on userspace and is in no way related to dkms, in fact they can even move to Fuse and still require dkms because the code is out of tree.

                I think what you wish is for Zol to be integrated on the main tree but i dont think that could happen license wise until Oracle stop being asshats
                yup, I would like it to be in kernel, I think some (ubuntu?) moving it to...

                Originally posted by starshipeleven View Post
                what's wrong with dkms?
                Nothing, it just me I just don't like it to be there, I understand why it is there and what it is for, but still, there are some things you just do not like...

                For example: bad weather (like raining or -25C), but still those things are good and "healthy" from other point of view

                Comment


                • #9
                  Does anyone know what they mean by:
                  Code:
                  - Device removal
                  I have a zpool that is messed up right now, due to the fact that while trying to replace 4TB drives with 8TB drives, I accidentally added one of the 8TB drives as a separate top-level vdev in the pool. When I couldn't get it to replace one of the 4TB drives with the one I just added, I erased the front part of the 8TB drive in an effort to make it ignore it so I could try again. After doing some searching, this seems to be a huge no-no in ZFS (which to me seems like a serious design flawin ZFS, but that's a rant I'll save for another day). Anyway, if this new feature would help me remove the missing device and bring the zpool back to life (I never wrote anything to the pool after adding the device), that would be awesome!

                  Comment


                  • #10
                    Originally posted by Turtleggjp View Post
                    Does anyone know what they mean by:
                    Code:
                    - Device removal
                    I have a zpool that is messed up right now, due to the fact that while trying to replace 4TB drives with 8TB drives, I accidentally added one of the 8TB drives as a separate top-level vdev in the pool. When I couldn't get it to replace one of the 4TB drives with the one I just added, I erased the front part of the 8TB drive in an effort to make it ignore it so I could try again. After doing some searching, this seems to be a huge no-no in ZFS (which to me seems like a serious design flawin ZFS, but that's a rant I'll save for another day). Anyway, if this new feature would help me remove the missing device and bring the zpool back to life (I never wrote anything to the pool after adding the device), that would be awesome!
                    This Issue turned Feature talks about it and it links to the OpenZFS roadmap item.
                    Describe the problem you're observing Adding a device to a pool causes the device to be bound to the pool forever in most cases since removal of the device requires recreation of the pool without the device to be removed. While it's comm...

                    Comment

                    Working...
                    X