BTRFS on mdadm

1st Official Post
    • Official Post

    As a follow up to this thread it is no longer possible using the GUI to add/create a btrfs file system on an mdadm raid (software raid), selecting btrfs from file systems you get the following;


    Type -> BTRFS

    Profile -> Single, DUP, Raid0, Raid1, Raid10

    Devices -> self explanatory


    under devices if you select an mdadm raid you will see "At least two devices are required" if you select a raid option from the profile drop down


    The btrfs file system has moved to Profiles -> this site has some excellent information along with btrfs read the docs

    Raid is not a backup! Would you go skydiving without a parachute?


    OMV 8x amd64 running on an HP N54L Microserver

    • Official Post

    That isn't quite true. I just created a two disk mdadm mirror and then put a btrfs single on top. If you pick anything other than single (didn't try DUP) for btrfs, it makes sense it will require more than one "disk". You can't setup two raid arrays (one for mdadm and one for btrfs which would be top of mdadm) nor does it make sense to.

    omv 8.5.8-1 synchrony | 7.0 proxmox kernel

    plugins:

    omvextrasorg 8.0.3 | kvm 8.3.2 | compose 8.1.24 | cterm 8.0.3 | borgbackup 8.2.1 | tempmon 8.0.5 |

    mergerfs 8.0.2 | scripts 8.0.6 | writecache 8.2.4 | backup 8.1 | kernel 8.0.16


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


    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!

    • Official Post
    Quote from geaves

    That I understand, but from memory one could create an mdadm array then create/add btrfs on top, the addition here seems to be profiles, if single works then fine.

    Correct. Older versions of OMV didn't allow creation of btrfs raid/profiles. So, everything it created was a "single".

    omv 8.5.8-1 synchrony | 7.0 proxmox kernel

    plugins:

    omvextrasorg 8.0.3 | kvm 8.3.2 | compose 8.1.24 | cterm 8.0.3 | borgbackup 8.2.1 | tempmon 8.0.5 |

    mergerfs 8.0.2 | scripts 8.0.6 | writecache 8.2.4 | backup 8.1 | kernel 8.0.16


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


    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!

    Actually, I was the OP in the thread you linked. I had no problem creating a BTRFS file system on a MDADM raid array. The issue was trying to put LVM on top of an MDADM array. Starting with two completely blank and erased disks (no file system at all), I was able to create the array (md0) using MDADM, but then I couldn't use that array as the basis for a physical volume in LVM. All of my issues came before even trying to select or create a file system.

    • Official Post
    Quote from Louie1961

    The issue was trying to put LVM on top of an MDADM array. Starting with two completely blank and erased disks (no file system at all), I was able to create the array (md0) using MDADM, but then I couldn't use that array as the basis for a physical volume in LVM. All of my issues came before even trying to select or create a file system.

    So, I just took my btrfs on mdadm and deleted the filesystem from the web interface. I tried to create a PV from the mdadm array but I couldn't because it still had a btrfs signature on it. I ran wipefs -a /dev/md0 and then was able to create a PV from it. I then created a VG and LV and put a btrfs filesystem on the LV all from the web interface.

    omv 8.5.8-1 synchrony | 7.0 proxmox kernel

    plugins:

    omvextrasorg 8.0.3 | kvm 8.3.2 | compose 8.1.24 | cterm 8.0.3 | borgbackup 8.2.1 | tempmon 8.0.5 |

    mergerfs 8.0.2 | scripts 8.0.6 | writecache 8.2.4 | backup 8.1 | kernel 8.0.16


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


    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!

    Yeah, I couldn't do that, although I tried. I suspect it is because of the way I am passing through my drives to the OS in Proxmox.

    • Official Post
    Quote from Louie1961

    Yeah, I couldn't do that, although I tried. I suspect it is because of the way I am passing through my drives to the OS in Proxmox.

    Did you try running wipefs -a /dev/md0 (or whatever your array number is) in the guest?

    omv 8.5.8-1 synchrony | 7.0 proxmox kernel

    plugins:

    omvextrasorg 8.0.3 | kvm 8.3.2 | compose 8.1.24 | cterm 8.0.3 | borgbackup 8.2.1 | tempmon 8.0.5 |

    mergerfs 8.0.2 | scripts 8.0.6 | writecache 8.2.4 | backup 8.1 | kernel 8.0.16


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


    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 did not explicitly do that. I did use the Proxmox wipe disk function followed by the OMV "wipe" function in the gui.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!

Register Yourself Login