Posts by roccur

    I just tried to upgrade OpenMediaVault 3 + ZFS. It failed with errors, so I started over by reinstalling from openmediavault_3.0.36-amd64.iso and it failed again with the same errors:

    - updated everything.
    - uploaded openmediavault-omvextrasorg_latest_all3.deb
    - installed omv-extras under "plugins"
    - enabled OMV-Extras -> ZFS + ZFS Testing
    - Update everything again
    - then installed openmediavault-zfs 3.0.4 under plugins which threw the same errors when trying to upgrade. Here is the part that shows what is going wrong:

    Here is a part from /var/lib/dkms/spl/ where I saw errors:

    Here are my versions:

    How can I fix this?



    Picking up on this: Thanks for the information everyone provided on this issue so far. Working on mounting openmediavault 3.0.46 NFS shares from various clients that use NFS version 3 I have the same issue but have a restriction so I can't add "/exports/"

    Via nfs version 4 it works nicely:

    $ showmount -e nas
    /export/logs *

    and logs can be mounted this way:
    $ mount nas:/logs /mnt/test123/tmp

    But an NFS version 3 client cannot:

    $ mount nas:/logs /var/tmp/test
    nfs mount: nas:/logs: Permission denied

    Yes, it works by inserting /export/ as you guys said in the thread:

    $ mount nas:/export/logs /var/tmp/test

    but I must get it to work with the short path on NFS version 3 because of a large number of NFS version 3 clients that use automount (which inserts the 'export' if it's there) in combination with programs that use a hardcoded NFS path.

    Question: How can offer the short path to all clients? I found an ugly "hack" (see below) but would like to do it properly in the GUI.

    the ugly workaround is:

    on OMV:
    $ mkdir /logs
    mount this new folder in /etc/fstab:
    /rz2pool/logs/logs /logs none bind 0 0
    add the new folder to /etc/exports:
    /logs *(rw,subtree_check,secure,no_root_squash)
    (this hack survives a reboot)


    Just wanted to provide an update about what I ended up doing: I used ZFS to create a single raid-z2 pool made up of 2 vdevs of 8 drives each ( by first creating a pool of 8 drives in the OMV GUI and then selecting "Exapand Pool" and adding the next 8 drives). The first 9 TB have been copied to new file systems created in the pool.
    Thanks @ryecoaaron and @hoppel118 for advice.

    Fyi, here are my versions:

    dpkg -l |grep zfs
    ii debian-zfs 7~jessie amd64 Native ZFS filesystem metapackage for Debian.
    ii libzfs2 amd64 Native ZFS filesystem library for Linux
    ii openmediavault-zfs amd64 OpenMediaVault plugin for ZFS
    ii zfs-dkms all Native ZFS filesystem kernel modules for Linux
    ii zfsutils amd64 command-line tools to manage ZFS filesystems
    dpkg -l |grep openm
    ii openmediavault 3.0.46 all Open network attached storage solution
    ii openmediavault-keyring 1.0 all GnuPG archive keys of the OpenMediaVault archive
    ii openmediavault-omvextrasorg 3.3.3 all Package Repositories for OpenMediaVault
    ii openmediavault-zfs amd64 OpenMediaVault plugin for ZFS

    I did give you a tip. Don't use software raid and use hardware raid. Definitely don't put software raid on top of software raid. You will regret that. zfs might be ok but not mdadm. If you can afford all those 8tb drives and space to back them up to, a hardware raid card is a small cost.

    Thanks, got it now :) . I will consider getting controllers but the system is already over budget and it's not only up to me. As an alternative, what would you think about 2 separate Raid6 software arrays if I'd be willing to get 2 separate file systems out of it?

    Report your bugs or issues with the omv-zfs plugin in the following thread:

    Maybe someone there can help you. ;)

    Greetings Hoppel

    Thanks, will do if I go down the ZFS route!

    @ryecoaaron Thanks, yes, redundancy is needed (2 failed drives should be tolerated). Backup is not an issue, it will happen continuously via rsync.

    I'd like to work with what I have (or is it such a disaster that it's not possible without getting problems later on? Things look like they are working so far). Given that, I'd appreciate if you could give me a tip regarding the RAID questions, even if a hardware raid controller would be more suited for the job.

    @hoppel118 I have considered ZFS under Freenas with the exact config you mention, but ran into an issue with not being able to replicate NFS paths as they were before (Freenas requires /mnt/data/...). But ZFS under OMV may be different - I will check out if I can get around the issue under OMV.


    I want to create a single large volume from 16 x 8TB drives using openmediavault_3.0.36 (downloaded the iso) on a system with 16 x 8TB drives connected to an ASRock X99 Extreme 11 motherboard.

    What I have done:
    -created 2 RAID6 arrays (each one consisting of 8 drives) since I have heard it's not good to have too many drives in a single array.
    -now there are 2 devices but to have a single file system you need 1 device. So I have combined the 2 RAID6 arrays into a single array using RAID0 (stripe) on top and it looks like it worked.

    Is there a better way to create a large volume from my 16 drives? (I can still start over at this point)
    Is RAID0 on top of RAID6 ok in general? I want to avoid complications in the future and nested arrays may be bad.
    Is it correct that 2 RAID6 with 8 drives is better than 1 RAID6 with 16?

    Thanks in advance!