mergerfs pool only works with manual restart

  • I have had a mergerfs pool since OMV5 (now OMV6) all working fine (rpi4 with 4 USB drives).

    Since upgrade to OMV6 though, there is some strange behaviour:

    After every reboot, the mergerfs pool is not available in File Systems.

    I then have to go into Mergerfs Plugin and manually click "Restart pool", which gives a browser error (see below).

    After that mergerfs is available in File Systems and I need to restart Docker and all works well.



    ipdNM4.jpg

    • Official Post

    Didn't we discuss this in your other thread already?

    omv 7.1.1-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.14 | compose 7.2.1 | k8s 7.1.1-1 | cputemp 7.0.2 | mergerfs 7.0.5 | scripts 7.0.8


    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

    we never resolved it

    If I can't reproduce it or figure out why from your info, I don't know what changes to make. Who else do you think is going to help/make changes? I'm not sure what starting a new thread will do to help.

    omv 7.1.1-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.14 | compose 7.2.1 | k8s 7.1.1-1 | cputemp 7.0.2 | mergerfs 7.0.5 | scripts 7.0.8


    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!

  • Other users.

    I run 2 rpi based servers and both show exactly the same issue.

    I assume there are other Raspberry users who come across the same issue and I am curious what their findings are.

    Isnt that the purpose of the forum?

    OMV6 HP t630

    OMV6 Xeon / i5 - SCSI PC

    OMV6 on Raspberry Pi4

    OMV5 on ProLiant N54L (AMD)

    • Official Post

    Other users.

    I run 2 rpi based servers and both show exactly the same issue.

    I assume there are other Raspberry users who come across the same issue and I am curious what their findings are.

    Isnt that the purpose of the forum?

    Sure but we had all of the info/questions/investigation notes in the other thread.


    I'm glad you have two that show the same issue but that doesn't help me. The mergerfs plugin has zero issues on my sata connected systems. I don't have a bunch of usb drives (none that are hard disks) and a lot of time to reboot a system waiting for failures. I know you want to blame me for changing from fstab to systemd but all fstab entries are converted to systemd.


    If you and other users can figure it out, I will gladly make changes to the plugins.

    omv 7.1.1-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.14 | compose 7.2.1 | k8s 7.1.1-1 | cputemp 7.0.2 | mergerfs 7.0.5 | scripts 7.0.8


    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 am not blaming anyone as I assume that systemd is the better solution (although from a user perspective it makes things a lot more complicated to maintain). Nevertheless, I am sure there are many advantages to scrapping fstab and I am just trying to fix this little RPI issue. The USB drives are pretty fast (all SSDs with USB3 link) and there are no delays in the boot process.

    So I think it may just be the order in which some boot steps are taking place or some timing overlap.

    So perhaps someone has seen the same issue....

    Is there a way to insert delays into the boot sequence? I am happy to try a few things.

    OMV6 HP t630

    OMV6 Xeon / i5 - SCSI PC

    OMV6 on Raspberry Pi4

    OMV5 on ProLiant N54L (AMD)

    • Official Post

    Is there a way to insert delays into the boot sequence?

    Sure but then any service that is waiting on the mergerfs pool (like docker for example) would fail. I don't consider this a solution though.


    sudo systemctl edit srv-mergerfs-POOLNAME.mount


    and add:

    Code
    [Service]
    ExecStartPre=/bin/sleep 30

    omv 7.1.1-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.14 | compose 7.2.1 | k8s 7.1.1-1 | cputemp 7.0.2 | mergerfs 7.0.5 | scripts 7.0.8


    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!

  • Docker fails currently anyway so maybe we should delay docker and trigger Pool Restart at the end of the boot sequence.

    is there a command to Restart the Pool?

    OMV6 HP t630

    OMV6 Xeon / i5 - SCSI PC

    OMV6 on Raspberry Pi4

    OMV5 on ProLiant N54L (AMD)

    • Official Post

    probably some creative way to do that but I cant think of a way right now. might as well add longer delay to docker.

    omv 7.1.1-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.14 | compose 7.2.1 | k8s 7.1.1-1 | cputemp 7.0.2 | mergerfs 7.0.5 | scripts 7.0.8


    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!

  • One of the 2 rpi servers has LUKS drives but they are decrypted automatically at boot. The other server has no LUKS.

    The pool error is the same for both.

    OMV6 HP t630

    OMV6 Xeon / i5 - SCSI PC

    OMV6 on Raspberry Pi4

    OMV5 on ProLiant N54L (AMD)

Participate now!

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