RAID 5 system degraded already

  • Only installed OMV last night on a HP N40L microserver. I have a 250GB disk on which I installed OMV and 3x500GB disks that I built a RAID5 array with. The disks weren't new, but as far as I was aware, they were good.


    The build process seemed to go quite smoothly and took a few hours to complete.


    Last night I used the USB Backup plugin to copy the contents of an external USB drive to a share - about 400GB of data.


    I am just trying to get my head around moving data around effectively, and I have noticed that my RAID 5 array is now degraded. Only 3 disks in total are visible in the Physical Disks view - my 250GB OS drive and 2 of the 3 array drives. Where is the third? If it had failed, I would at least expect it to show up in here somewhere...

    • Offizieller Beitrag

    The initial sync would be hard on an older drive. What's the output as root:


    fdisk -l
    cat /proc/mdstat

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • Please bear with me as I am very new to all of this.


    In the console I typed in
    fdisk -l
    and clicked return. I am faced with all sorts of figures, so I am not sure what info to give you. A couple of things that stand out are as follows:-


    Disk /dev/sdc doesn't contain a valid partition table
    Disk /dev/sdd doesn't contain a valid partition table
    Disk /dev/md127 doesn't contain a valid partition table


    There is also other information that has scrolled off my screen.


    Is this any use?

    • Offizieller Beitrag

    Did the fdisk command list a /dev/sda and /dev/sdb? If so, try this as root:


    mdadm --stop /dev/md0
    mdadm --remove /dev/md0
    mdadm --zero-superblock /dev/sdd
    mdadm --add /dev/md127 /dev/sdd

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • When I logged into the web console, I noticed that it was sdd that was missing. I think sda is the OS drive, and my raid was built of sdb, sdc and sdd.


    When I look at my N40L server, there are 4 drive slots. The left most drive is the 250GB OS drive (sda). I therefore assumed that the second slot must be sdb, the third slot sdc and the last slot sdd.


    Not wanting to hang around, I pulled out the fourth drive and replaced it with the only other drive I had lying around, which is a 1TB drive. Rebooting and messing around with the web admin, I seem to have managed to tell the system to rebuild the array using the new drive.


    Its been in there about 20 minutes now and is still showing that the rebuild is at 0%. I hope this works. Not the greatest of starts in my world of OMV.

    • Offizieller Beitrag

    sdd showed up in fdisk which is more important than showing up in the web interface. I'm guessing you didn't run the commands? And yes, rebuilding the raid can be slow.


    Maybe not the greatest of starts but you are new to Linux and using advanced features. Throwing older hardware in the mix is not helping. A little patience and you will have a very nice system :)

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • No sorry, didn't see the commands you suggested before I had pulled the disk out.


    Thanks for your words of encouragement - much appreciated.


    I am a little worried that the sytem has allegedly been rebuilding for over 45 minutes and it is still at 0%. Do you think I have an issue here?

    • Offizieller Beitrag

    That seems slow. The output of cat /proc/mdstat will tell you the rate it is rebuilding at. You could try the following command as well:


    echo 50000 > /proc/sys/dev/raid/speed_limit_min


    It defaults to 10000

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • having typed
    cat /proc/mdstat it seems it has completed 0.4% and speed is 393k/sec!!!!


    Rebooted and chose the 'safe' mode. Now reading 1500k/sec. At that speed it thinks it will take nearly 4 days to rebuild. Does that seem about right to you?

    • Offizieller Beitrag

    That is very slow. Type tail /var/log/messages and see if you see any errors.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • Checking again and it is now at 201k/sec


    What could be causing things to be so slow?


    Just typed tail /var/log/messages and nothing is jumping out as an error. A few references to usb 5-1, but that might just be a reading of available usb ports I guess?

  • When my machine reboots, I do see a few red ERROR/FAILED messages pop up but it is so quick I don't get a chance to see what they are.

    • Offizieller Beitrag

    Hard to say. If you were trying to use the raid array, it would definitely slow it down. Maybe you have a bad cable?

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • Rebooted in normal mode and it seems to be running at about 1400k/sec. Still really slow so I am questioning the advantages of RAID in my situation. With hardware RAID, I imagine you get much more speed, but this is software raid.


    Am I flogging a dead horse? Should I just reinstall the N40L with a license of Windows Home Server 2011 and use a combination of regular external backups and online, offsite backups (aka Crashplan)?

    • Offizieller Beitrag

    Software raid can be very fast. Mine is. I routinely transfer files from a virtual machine on my server to another location on the server at over 160mb/s.


    I personally wouldn't use Home Server but I am a big linux fan. I'm not sure you need raid. You could use one drive for music and one for pics. Then you could schedule backup using rsync to copy to the third drive. It would only copy the change content making it very fast. And then, you could use external and offsite backups. Greyhole would really fit your need but you would need to stick with omv 0.3 to use it (that isn't as bad as it sounds).

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

    • Offizieller Beitrag

    You have to delete the shares in samba/cifs,then the shared folders, and unmount the filesystem. Then you can use the Wipe button on the Physical Disks tab.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    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!

  • Managed to delete the array and rebuilt 3 separate drives with different shares on each.
    500GB for Photos
    500GB for music
    1TB for backup (for when I learn how to use rsync)


    Is there a way to test the integrity of my hard disks within OMV? I know there is the SMART monitoring, but I am not really sure what to do with that or indeed if it is the right thing to be using.

Jetzt mitmachen!

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