OMV 5 boots in emergency mode, cannot access web GUI

  • Hello dear OMV folks,


    I am a complete newbie to both Linux and OMV, so please use short words and long sentences. I looked through the forum for similar issues, but after going through the proposed solutions and encountering yet more problems, I decided to start a new thread.


    I made an OMV-based NAS to have a secure location to store photos of my newborn daughter. I installed OMV 5.6.13 on a 128 GB Verbatim SSD drive and use two 4TB Seagate IronWolf disks in a RAID1 array for data storage. All disks were purchased new, are around 7 months old now. I also installed a Plex docker/portainer and have been thrilled with the results so far. The system has been working for around 6 months without a glitch, I only needed to restart the Plex container every once in a while.


    In mid-August, we left for a holiday and I decided to turn the NAS off. After coming back last week, I wanted to upload the photos but could not access the web interface. I connected a screen and a keyboard to the NAS and saw the following message:


    Code
    [FAILED] Failed to start File System Check on [one of the two 4TB drives]. 
    See 'systemctl status...
    [DEPEND] Dependency failed for [same 4TB drive]
    [...] some more linuxy stuff
    You are in emergency mode. After logging in, type "journalctl -xb" to view system logs, "systemctl reboot" to reboot, "systemctl default" or "exit" to boot into default mode.
    Give root password for maintenance (or press Control-D to continue):


    The system logs were not very helpful to me as they displayed 15+ screens of stuff I do not understand. After finding this thread, I tried to do what Zoki said:

    Zitat

    Log in,

    mount -o remount,rw /

    comment out the mergerfs from /etc/fstab

    reboot


    I did not do anything to /etc/fstab, but it seemed to do the trick. The system started, the web GUI connected, I mounted the file system, the shared folders appeared in the Windows network and all looked normal. At this point, the NAS was still gutted on my desk, so I happily turned it off, put it back together and placed it where it normally goes. After booting it, I was again in emergency mode.


    I went back to the fstab and commented out all three drives. The system started, but the web GUI did not connect, neither did PuTTY. I uncommented the UUID of the SSD drive with OMV. Now I was able to connect via the web GUI, but could not mount the file system. So I uncommented the 4TB drives, leaving the fstab in its initial state. However, now the dashboard shows me that the SMB/CIFS is Enabled but not Running (red dot under Running) and I cannot connect to the shared folders.


    Any help would be much appreciated... especially help which does not cause me to wipe the data from the storage drives ;)


    EDIT: I disabled and re-enabled the SMB/CIFS and now seem to have access to all files. Plex ist not connecting but I can live with this for now. I would appreciate any suggestions on how to prevent this from happening in the future.

    • Offizieller Beitrag

    my newborn daughter

    Congratulations

    I installed OMV 5.6.13

    OMV5 is end of life, current version is OMV6. You should upgrade as soon as possible.

    in a RAID1

    Why RAID1? You know that RAID is not Backup?

    The system logs were not very helpful to me

    Maybe the not maimed logs were helpful to others here on the forum.

    especially help which does not cause me to wipe the data from the storage drives

    Do you have backup?


    Unfortunately your system is in a unknown state now. You did changes to configuration files. Do you have a recent backup of your OS filesystem?

  • Hi Macom,


    Thank you for your quick answer. I have downloaded the OMV6.0.24 iso and will upgrade one of these days. I hope that, being on separate drives, my files will remain unspoilt. I am currently backing up the photos, the movies on Plex are a secondary concern. Maybe I will opt for a NTFS file system to have some more flexibility. One thing I did not mention in my original post is that I plugged the two drives into a Windows machine, but the Linux Reader I downloaded only showed unallocated space where I was expecting to see ext4. So the files were only accessible through OMV.


    There are two file systems on the SSD:

    /dev/sda1 vfat 148kb of 510 MB used

    /dev/sda2 ext4 8 GB of 115 GB used

    So definitely not full... if the root file system is a different one, I don't know how to check it.


    I chose the RAID1 mirroring to have a second copy of the files if one of the drives should fail. Can you recommend a drive diagnostic tool?

  • Hi donh, thank you for the tip. Below is a table of what percentage is used, it does not look like a lot to me...


    FilesystemSizeUsedAvailUse%Mounted on
    udev
    3.7G03.7G0%/dev
    tmpfs757M17M740M3%/run
    /dev/sda2116G8.0G102G8%/
    tmpfs3.7G03.7G0%/dev/shm
    tmpfs5.0M05.0M0%/run/lock
    tmpfs3.7G03.7G0%/sys/fs/cgroup
    tmpfs3.7G12K3.7G1%tmp
    folder2ram3.7G23M3.7G1%/var/log
    folder2ram3.7G03.7G0%/var/tmp
    folder2ram3.7G1.2M3.7G1%/var/lib/openmediavault/rrd
    folder2ram3.7G20K3.7G1%/var/spool
    folder2ram3.7G15M3.7G1%/var/lib/rrdcached
    folder2ram3.7G4.0K3.7G1%/var/monit
    folder2ram3.7G1.3M3.7G1%/var/cache/samba
    /dev/md03.6T727G2.9T20%[uuid of storage disk from original post]
    /dev/sda1511M148K511M1%[uuid of SSD]
  • An update on the whole boots in emergency mode situation. I am pretty sure I made lots of mistakes, but here goes:


    After saving all the data I needed, I decided I could not risk an even worse failure. I installed OMV 6.0.24 without a hitch, it even recognised my RAID array as such and was able to mount the file system and folders the way I had them before. So all good right? Well, not so much. When I went over to services, the system yelled at me that there is obsolete data cached. Not knowing how to clear the cache at the time, I rebooted the NAS system, whereupon I again was greeted by the failure and dependency messages from my original post. It was time for Scorched Earth. I wiped both 4TB drives, uncoupled the RAID array and set it again. This took about 18 hours of thumb twiddling. I created a new file system and tried to mount it, only to be shown a message that the structure needed cleaning, similar to this thread. Emboldened by the simple looking solution, I happily started PuTTY, logged in as root and entered


    fsck /dev/md0


    There were nearly 29000 blocks with wrong inodes to clean, so the whole thing took a while. Please do not ask how I confirmed every single question the system threw at me. Afterwards, I was able to mount my system just fine, only to stumble on the next thing. Or the previous thing. The services were still storing obsolete cached data. I then found this thread, with an embarrassingly simple solution. Well, you live and learn.


    Now my system is up and running, all photos have been copied and are accessible via the Windows Explorer, as well as from my Android phone (after some tweaks in the Windows settings). It is great for uploading the hundreds of baby photos this server was actually built for.


    In the end I did not really fix anything, just rebuilt from scratch. I still have no idea why the problem occurred in the first place. I guess I will have to set up a backup backup location. Applying  fsck  before I had wiped the drives might have remedied the situation... then again, it might have not. Now I go put The Lord of the Rings on my Plex again.

  • silverthorn

    Hat das Label gelöst hinzugefügt.
  • silverthorn

    Hat das Label OMV 5.x hinzugefügt.

Jetzt mitmachen!

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