Unable to mount HFS/HFS+ in OMV 4.1.7 for Pine64

  • Basically this. I've done apt-get install hfsplus hfsutils hfsprogs and even cat /proc/filesystems.
    The HFSPLUS system is still unknown type.
    the /proc/filesystems does not list HFS/+ in it.
    Is there any way to get these FS to mount as I'd like to use it for an incremental TimeMachine backup.

    • Offizieller Beitrag

    Basically this. I've done apt-get install hfsplus hfsutils hfsprogs and even cat /proc/filesystems.
    The HFSPLUS system is still unknown type.
    the /proc/filesystems does not list HFS/+ in it.
    Is there any way to get these FS to mount as I'd like to use it for an incremental TimeMachine backup.

    I would guess HFS isn't enabled in the kernel. grep ^CONFIG_HFS /boot/config-* will tell you if it is. If not, you would have to compile a new kernel to enable it.

    omv 7.0-32 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.9 | compose 7.0.9 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • I would guess HFS isn't enabled in the kernel. grep ^CONFIG_HFS /boot/config-* will tell you if it is. If not, you would have to compile a new kernel to enable it.

    Yeah, that returns nothing. Any advise on how to go about this process? Im 100% new to the Linux thing.


    Its worth noting that I was running your image originally, but am now getting started on the debian level.


    My other machine is an UBUNTU device. Is there any way to run OMV on UBUNTU? Its standard with HFS support.

    • Offizieller Beitrag

    Any advise on how to go about this process? Im 100% new to the Linux thing.

    Compiling a kernel is not something someone new to Linux wants to start doing.


    Its worth noting that I was running your image originally, but am now getting started on the debian level.

    The image is Debian built with Armbian scripts. Not sure what you mean.


    My other machine is an UBUNTU device. Is there any way to run OMV on UBUNTU? Its standard with HFS support.

    Nope. The fact that it is Ubuntu isn't the reason it has HFS support. Any kernel can have HFS support. It just happens that the armbian devs didn't enable it.


    I don't use timemachine but I'm not sure why the filesystem matters?

    omv 7.0-32 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.9 | compose 7.0.9 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Is there any way to get these FS to mount as I'd like to use it for an incremental TimeMachine backup

    Wrong approach. You don't need HFS+ support but POSIX compliance so simply choose between ext4 and btrfs for your data share.


    Netatalk will then share this filesystem via AFP (implementing the FPzzz and the needed sync calls) and OS X will then create a HFS+ sparsebundle on the network share (here the HFS+ semantics live).


    HFS+ support in Linux is more or less broken anyway but fortunately not needed with this use case. In general it's a very bad idea to move a disk that was connected to a client directly to an OMV host (regardless whether it's HFS+ and OS X or FAT32/ExFAT/NTFS and Windows).

  • Wrong approach. You don't need HFS+ support but POSIX compliance so simply choose between ext4 and btrfs for your data share.
    Netatalk will then share this filesystem via AFP (implementing the FPzzz and the needed sync calls) and OS X will then create a HFS+ sparsebundle on the network share (here the HFS+ semantics live).


    HFS+ support in Linux is more or less broken anyway but fortunately not needed with this use case. In general it's a very bad idea to move a disk that was connected to a client directly to an OMV host (regardless whether it's HFS+ and OS X or FAT32/ExFAT/NTFS and Windows).

    I thought it would be a clever idea to use HSF+ for my OMV installation because in case I have to copy large files to the disk I could remove it from the NAS and do it multiple times faster directly in OSX.
    It seems though that many people don't recommend to use HSF+ with a linux system.
    HSF+ on OMV seems to work for me, even with Journaling enabled. But sometimes the drives mount read only. That's odd.
    On the other hand I want to be able to use timemachine on the drives.


    May I summerize from your suggestion:
    ext4 over HFS+
    share via AFP in Mac environments, not SMB
    enable Timemachine with Netatalk

  • I thought it would be a clever idea to use HSF+ for my OMV installation because in case I have to copy large files to the disk I could remove it from the NAS and do it multiple times faster directly in OSX

    Please don't do this since this will introduce a bunch of problems you'll only realize once it's way too late and you already have a mess on these disks. Never share disks between clients and servers (unless the server is running OS X 10.9 or later -- there the file sharing daemons take care about the necessary translations). More details: Copy files internally via ssh allowed?


    I personally don't use ext4 on any data share since it's 2019 and we have ZFS and btrfs in the meantime (if you don't encrypt your TM backups the transparent filesystem compression feature of both can save you a lot of disk space, also server side snapshots provide a bit safety headroom wrt backup corruption). But of course you will be fine using ext4 or any other POSIX compliant filesystem and as long as you don't attach your server disks directly to any client.


    Currently (OMV up to version 4) you can use only AFP with TimeMachine, starting with OMV 5 it will be SMB instead.

  • Please don't do this since this will introduce a bunch of problems you'll only realize once it's way too late and you already have a mess on these disks. Never share disks between clients and servers (unless the server is running OS X 10.9 or later -- there the file sharing daemons take care about the necessary translations). More details: Copy files internally via ssh allowed?
    I personally don't use ext4 on any data share since it's 2019 and we have ZFS and btrfs in the meantime (if you don't encrypt your TM backups the transparent filesystem compression feature of both can save you a lot of disk space, also server side snapshots provide a bit safety headroom wrt backup corruption). But of course you will be fine using ext4 or any other POSIX compliant filesystem and as long as you don't attach your server disks directly to any client.


    Currently (OMV up to version 4) you can use only AFP with TimeMachine, starting with OMV 5 it will be SMB instead.

    Very well explained, thank you!


    Copy files internally via ssh allowed? is a very good guide!
    Not only does it explain in plain terms why it is vital to always use the client to access the storage via Filesharing (Samba, Netatalk),
    but it also describes the dangers of removing the disks from the server and mounting them on a mac/pc, messing around with the files directly.

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!