MergerFS - Internal Server Error

  • 500 - Internal Server Error when trying to create a new volume. attaching pics of the settings i'm using and the error i get after i click to apply the changes. Not sure where to look for logs to troubleshoot. Any suggestions?

  • Hopefully 4 weeks isn't too much of a necro. BUT I had a similar problem. It kept the drive from mounting and OMV booted in safe mode or whatever the proper term is. I managed to remount / rw and edit fstab to get booted with the wildcard method for my pools as discussed in the thread(s) linked above. I then had a drive in my drive pool (and snapraid collection) die on me.

    I managed to get all the data recovered to another drive in the collection from parity, resynced etc and everything seems happy now except the filesystems UI in OMV6. It's empty. I've managed to edit the config to take out the references to the now dead/removed drive, omv-salt deploy fstab gave me a good working fstab file.

    mergerfs still looks to have reference to the drive I removed from the pool (I'm using unionfilesystem plugin). I have no idea where to look at union/mergerfs config except fstab to know what's going on. fstab doesn't reference the dead drive in any way now so I'm lost.

    What I've done:

    1. I have edited omv's config.xml to remove the mntent (which was successful with the omv-salt deploy fstab results)
    2. I have edit config.xml to remove the drive from <snapraid> and the snapraid.conf was manually corrected (can I do a similar omv-salt deploy here to regenerate? it's not in the deploy list)
    3. I have also removed the <mntref>UUID-XXXXXX</mntref> from the /etc/openmediavault/config.xml under <unionfilesystems>

    To the questions:

    • How do I figure out what's causing the filesystems UI from doing its rendering? (dmesg and journalctrl and syslog are all silent on the issue) Disks UI is working fine
    • How can I "deploy" the config.xml settings which I believe are accurate/correct for
      • snapraid (even though it's working fine and doesn't seem to be needed
      • unionfilesystems+mergerfs (which still thinks the removed and now dead drive should be present)

    Ok, hopefully that's enough. I think the "route is too long" issue was the root problem and then it just kinda snowballed. I'm sure monkeying around in the config files wasn't the right solution but with nothing booting because of the bum fstab entry I was kinda stuck - I couldn't even load the UI to make corrections the "right" way.

Participate now!

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