Posts by Radikal

    I had issues with OMV 4.x install due to incompatability between my mobo and vanilla stretch. I needed to do an inplace upgrade from 8 to 9 and install kernel upgrades before reobooting to get around an issue similar to yours. Have a look at the in-place 9 to 10 Debian upgrade guide:


    https://www.debian.org/release…e-notes/ch-upgrading.html


    Take note of section 4.6 that talks about upgrading your kernel.


    (Don't try this on your live build - swap in a blank drive to test out a fresh install Stretch followed by an in-place Buster upgrade with kernel upgrades. Once this is sorted, you can install OMV on a stable OS.)

    OK, as OMV 3.x no longer had any updates, I didn't pursue the backports via that interface (I didn't realize the backports were not constrained). I'll run through a virtual machine build to see how this plays out for my own edification.


    Thanks


    -Ron

    Given my issue being unable to boot from a 4.x kernel lower than 4.16, I don’t think there was a point where I could reliably run OMV to do that. For this unusual installation conundrum, I’m booted up on Jessie running the 3.x kernel and have just done the 9.5 Stretch install (which installs the 4.9 kernel) but have not booted up on the 4.x kernel yet.


    I was operating under the notion that OMV expects that I would have rebooted into 4.x running the 4.x kernel prior to running the OMV install and using the OMV front end. Are you saying that I could have run the OMV install on top of the Debian 3.x to 4.x in-place upgrade without rebooting and I would have been able to use the OMV backports via the OMV front-end prior to that first boot into 4.x?

    n/a-backports? That is definitely wrong. Does omv-mkconf apt fix it?

    That appears to have done the trick:



    Code
    root@D8:~# cat /etc/apt/sources.list.d/openmediavault-kernel-backports.list
    deb http://httpredir.debian.org/debian stretch-backports main contrib non-free

    Update Manager no longer throws an error. No idea why that happened. I can recreate the problem on demand by following the above build sequence on a VM. Perhaps OMV’s build isn’t dealing with the stretch backports I had to add to get to 4.16 and above during the Debian install phase.

    2nd half:

    Here's the full error (too long to include in a single message):
    1st half:


    I'm getting the following error when I run Check in the OMV 4 Update Management screen:



    Note that my installation was on a Debian minimal install implemented via:

    • Fresh Debian 8.11
    • In place upgrade to Debian 9.5
    • In place backport kernel 4.17 added
    • OMV installed per OMV installation docs when installing onto Debian

    Here's my sources list and my OMV sources lists:

    It appears that the OMV kernel backports sources.list is giving me grief. Suggestions?


    -Ron

    After running all the OMV updates, I appear to be in a good place.


    I do get warnings against the stretch backports source and the weak references error.


    I’m looking into whether or not the backports source is still needed as normal maintenance should have taken me to 4.17 (in which case I can comment that out from the sources file - I think).


    I’m also sitting tight awaiting a maintenance fix for the weak references script error.


    So, I’m mostly back in business with SMB and backups working.

    I landed on the following sequence:

    • Netinst Debian 8.11 ISO for amd64
    • At conclusion of install, reboot and log in as root
    • Do an in-place Debian 8 to 9 upgrade per Debian instructions
    • At conclusion I have a 3.16 kernel and a newly installed 4.9 kernel
      • I know that I can’t successfully boot from the 4.9 kernel
    • Prior to rebooting, I add the stretch backports to sources and install the 4.17 kernel
    • At conclusion, I reboot and have to add “noacpi” to the 4.17 kernel for it to find the boot drive
    • Now I install OMV 4 on Debian 9 stretch per the instructions including this command:


    Fails due to —allow-downgrades and —allow-change-held-packages not being recognized. I ran w/out those but expect there are issues. OMV starts up but the net stats portion of the dashboard generates a script failure pop up.

    That is exactly what I used. Debian 9 netinst ISO boots from a generic kernel and installs the 4.9 amd64 kernel. What I’m now looking for is one of these paths forward:

    • Find a Debian 9 netinst ISO that installs a 4.16 or later kernel
    • Find a way to upgrade the kernel at the conclusion of the netinst process prior to reboot
    • Find a utility that I can boot into to do the kernel upgrade after the installation completes

    Any ideas?

    OK, searching the OMV forums I see that manually installing omv extras from the command line fixes this error.


    I was then able to install the updates through to 4.17 kernel and everything was working in the VM


    So, I proceeded to do a live build on my problem machine.


    Built and booted from the Debian 9.5 ISO (boots the 4.9 Kernel)
    Keyboard wasn't recognized but a warm boot got me past that.
    Proceeded with same install steps without issue.
    On reboot, the 4.9 Kernel also fails but doesn't trigger a panic:



    So, I need to figure out how to upgrade from 4.9 to at least 4.16 for the kernel to successfully boot. Any ideas on an alternate built to install or a way to upgrade this one after the install but before rebooting the first time?


    -Ron

    Working through this on a virtual machine first (Oracle VM VirtualBox)...


    • Booted from Debian 9.5 amd64 ISO and installed the OS

      • Took all the defaults


      • Had to note which disk (sda..sdd) was my boot disk as the grub installer's prompt is not helpful
      • Only installed the system utilities, skipped the printer and desktop
    • Executed the manual install steps from the OMV docs (skipped the comment lines as this was manually typed in) from root


      • export LANG=C
        export DEBIAN_FRONTEND=noninteractive
        export APT_LISTCHANGES_FRONTEND=none
        apt-get update
        apt-get --allow-unauthenticated install openmediavault-keyring
        apt-get update
        apt-get --yes --auto-remove --show-upgraded \
        --allow-downgrades --allow-change-held-packages \
        --no-install-recommends \
        --option Dpkg::Options::="--force-confdef" \
        --option DPkg::Options::="--force-confold" \
        install postfix openmediavault
        omv-initsystem


    • Successfully connected to the OMV web front-end
    • Received an error message on the Update Management and Plugins screens:

      • Error
        The index of upgradeable packages does not exist. Please re-synchronize the package index files from their sources.
        OK Show Details
      • Details are:
      • Error #0:
        OMV\Exception: The index of upgradeable packages does not exist. Please re-synchronize the package index files from their sources. in /usr/share/openmediavault/engined/rpc/apt.inc:114
        Stack trace:
        #0 [internal function]: OMVRpcServiceApt->enumerateUpgraded(Array, Array)
        #1 /usr/share/php/openmediavault/rpc/serviceabstract.inc(123): call_user_func_array(Array, Array)
        #2 /usr/share/php/openmediavault/rpc/rpc.inc(86): OMV\Rpc\ServiceAbstract->callMethod('enumerateUpgrad...', Array, Array)
        #3 /usr/sbin/omv-engined(536): OMV\Rpc\Rpc::call('Apt', 'enumerateUpgrad...', Array, Array, 1)
        #4 {main}

    I restarted the virtual machine (reboot) but that did not correct the issue. How does one re-synchronize the package index files from their sources so I can apply updates?

    I've been working my way through that path as prompted by the formal OMV docs. So far, it looks like I should do the following:



    I'll update with where I've gotten to and any further questions. Thanks

    Had been running OMV 2.2. Win 10 SMB v1 change prompted me to finally upgrade. Install of OMV 4.1.3 from USB uneventful until the first post-installation reboot. Receive:


    Edited the boot line as instructed with noapic but still no joy.


    Tried installing OMV 3.0.94. That installs and boots fine but, as it is now EOL, can't update it nor do an in-line upgrade, so this isn't a path forward.


    Found a reference to this scenario on the Debian.org site: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=883294
    They have the same motherboard and also experience the same symptoms.
    They tried booting with "no_timer_check" as advised and that doesn't work either.


    They recommend installing the 4.16 or later kernel and that was verified by the user.


    So, is there an OMV 4.x build using a 4.16 or later kernel I can use? I'm dead in the water until I can rebuild the system to reaccess my stuff on the data drive.
    :/
    -Ron

    In tekkb's guide "Headless Backup of the System Drive with Clonezilla", can I target an openmediavault raid drive for the image and be able to restore from it or do I need to setup a non-raid drive for storing headless backups?



    -Ron

    Basic navigation and file manipulation is essential when troubleshooting. So:


    Navigating the directory tree and Linux specific device structure (where's my data?)
    Searching for files (including how to see pertinent attributes)
    Editing files (with good backout practices)
    Basic concepts on installing and maintaining things - a common point of confusion
    Doing simple tips - adding shortcut commands with useful switches, simple boot time automations


    Also point them to how to backup and restore their build before experimenting with your tutorial ;-)
    -Ron