Mounting through Web Interface throws error and lets OMV reboot

  • Hi community,


    I just added a new software raid and waited till it got completely synced. Afterwards I created an ext4 filesystem. As soon as this has been finished I wanted to mount the filesystem. This gave me an error in the web interface (502 - bad gateway). My server rebooted itself (slowly) afterwards. Slowly means it did not disappear immediatly, but slowly shut down and rebooted.


    After the reboot the web interface showed me a pending change. When I tried to apply it now, it rebooted again. When I dismissed it, everything was fine, except it was not mounted. I could not find anything suspicious in the syslog.


    When I use salt to deploy fstab, the drive is mounted, but I cannot add a shared folder.


    Any ideas?

  • When I tried to apply it now, it rebooted again.

    Openmediavault do not do an automatic shutdown or reboot. Which hardware platform do you use? It sounds like a problem of the power supply. For me this could be hardware related.

    OMV 3.0.100 (Gray style)

    ASRock Rack C2550D4I C0-stepping - 16GB ECC - 6x WD RED 3TB (ZFS 2x3 Striped RaidZ1) - Fractal Design Node 304 -

    3x WD80EMAZ Snapraid / MergerFS-pool via eSATA - 4-Bay ICYCube MB561U3S-4S with fan-mod

  • I think something on the OS side fails. This is a reproducable behaviour. I have an Intel 1155 Sockt board with a Xeon on it. It works since several months and is reliable over month, except I do what is described above. I even could reproduce it on different hardware. Maybe the amount of drives is too much (9 drives at all)?

    • Offizieller Beitrag

    Maybe the amount of drives is too much (9 drives at all)?

    I've had 20+ drives connected to an OMV system with no problem before. I would check syslog and/or the screen to see why it is rebooting.

    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!

  • After the reboot when I apply the pending config changes, the watchdog sends me following mail:

    Message from watchdog:

    The system will be rebooted because of error 2 = 'No such file or directory'


    And this is the syslog (between 19:32 and 19:34):



    Sep 25 19:32:02 blackbox systemd[24529]: Reached target Sockets. Sep 25 19:32:02 blackbox systemd[24529]: Reached target Basic System.

    Sep 25 19:32:02 blackbox systemd[24529]: Reached target Default. Sep 25 19:32:02 blackbox systemd[24529]: Startup finished in 56ms.

    Sep 25 19:32:02 blackbox systemd[1]: Started User Manager for UID 0. Sep 25 19:32:02 blackbox systemd[1]: Started Session 67 of user root. Sep 25 19:32:37 blackbox kernel: [ 7135.772237] watchdog: watchdog0: watchdog did not stop! Sep 25 19:34:49 blackbox systemd-modules-load[351]: Inserted module 'coretemp' Sep 25 19:34:49 blackbox kernel: [ 0.000000] microcode: microcode updated early to revision 0x21, date = 2019-02-13 Sep 25 19:34:49 blackbox kernel: [ 0.000000] Linux version 5.7.0-0.bpo.2-amd64 (debian-kernel@lists.debian.org) (gcc version 8.3.0 (Debian 8.3.0-6), GNU ld (GNU Binutils for Debian) 2.31.1) #1 SMP Debian 5.7.10-1~bpo10+1 (2020-07-30) Sep 25 19:34:49 blackbox kernel: [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-5.7.0-0.bpo.2-amd64


    Einmal editiert, zuletzt von MrFloppy () aus folgendem Grund: Added syslog

    • Offizieller Beitrag

    Something is going on with the system according to watchdog. So, it is unresponsive. What kernel are you using?

    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!

  • Yes, but unfortunately I cannot see what exactly.


    5.7.0-0.bpo.2-amd64 #1 SMP Debian 5.7.10-1~bpo10+1 (2020-07-30) x86_64


    This is the kernel I use.


    I had the same error before, several times. Usually using salt to fix fstab solved the issue. This time unfortunately not.

    • Offizieller Beitrag

    Try the proxmox kernel (tab to install it in omv-extras) to see if it helps. And is there anything in /var/log/watchdog/?

    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!

  • Thanks for your suggestions.


    The watchdog log directory is empty and I trying the proxmox kernel now. I will update this answer afterwards.


    One thing I think I saw: It looks like OMV is unmounting the "/" (root) partition, when trying to mount a new partition, which triggers something to restart the whole machine. But this is just a guess.


    UPDATE: No change in behaviour with the proxmox kernel. When clicking "Mount" I still get "

    Removing the directory '/' has been aborted, the resource is busy." followed by a crash of the Web-UI and then a reboot.

  • Anyone else has suggestions or knows what is happening when I press the mount button?

    I do not understand how it can happen, that the root Filesystem seems to get unmounted.


    Or is there a possibility to see pending changes and what they look like?


    Thanks a lot.

  • As an update:
    I added the drives (after formatting them through the Web-UI) manually to OMV's config.xml.

    Afterwards I did a "omv-salt deploy run fstab" and the drives work as expected. I just needed to add them to collectd and everything is as expected as of now, while the mounting through the web interface still did not work, which makes it look like a bug in OMV to me.

  • A further update:
    Using remote-mount also makes the system reboot, as soon as I press the button "mount".

    Fortunately, this time I got an error in the syslog:


    Code
    smartctl[30802]: segfault at 0 ip 00007f796261df5b sp 00007fffed048880 error 4 in libc-2.28.so[7f79625d0000+148000]


    I tried reinstalling smartmontools with the version from buster-backports. This removed the error, but did not change the behaviour...


    UPDATE: In the kernel log the last entry before every reboot is as follows:

    Code
    Mar 17 16:46:17 blackbox kernel: [   57.447690] softdog: initialized. soft_noboot=0 soft_margin=60 sec soft_panic=0 (nowayout=0)
    Mar 17 16:46:17 blackbox kernel: [   57.447692] softdog:              soft_reboot_cmd=<not set> soft_active_on_boot=0

    I am digging deeper.


    Another update:


    The watchdog gets triggered when changing the mount config (probably fstab). I stopped the watchdog and the system is not rebooting any more. Nevertheless, the system is unusable, as no mounts are present. I will continue checking.



    Any ideas for error research?

    4 Mal editiert, zuletzt von MrFloppy () aus folgendem Grund: Another update on the watchdog

  • Yep, similar story for me: had a phantom file system with mount point /. Trying to delete it through the GUI threw error. Removed it from config.xml, which made it go away.


    The next FS I created throws the "error unmounting /", followed by "bad gateway" every 10 secs. As soon as I OK the first error the system reboots, then asks me to confirm changes. If I confirm it goes back into the "bad gateway" loop, if I revert everything looks good, but can't mount my new FS.


    I've created my own topic for this, but just wanted to put that here to confirm that there seems to be a bug.

Jetzt mitmachen!

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