Posts by Griffo

    OK so I started again, and this time it seemed to go mostly fine. In case any other idiot does this, here's how I got it running


    apt-get remove --autoremove openmediavault

    Then instructions from https://openmediavault.readthe…stallation/on_debian.html


    Then omv-engined would not auto-start so..

    systemctl unmask openmediavault-engined


    OMV-Extras would not re-install until I did a

    apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 7AA630A1EDEE7D73


    And apart from the fact that it seems to now cache my login in the browser (if i reboot it, it does not ask me to re-auth) it seems to all work fine.

    FIngers crossed.

    Thanks. I got various errors trying that, so at this point i'm just trying to get a completely clean OMV install. I'm OK with re-doing all the OMV config, but want to avoid re-doing the entire server as I have other stuff (HDHR, VIrtualBox, Unifi) running on the box


    I've done:

    apt remove openmediavault

    apt autoremove

    apt clean


    Then followed the steps at https://openmediavault.readthe…stallation/on_debian.html


    I get the console, but then when I try to log in I get the following:


    Code
    Error #0:
    OMV\Rpc\Exception: Failed to connect to socket: No such file or directory in /usr/share/php/openmediavault/rpc/rpc.inc:141
    Stack trace:
    #0 /var/www/openmediavault/rpc/session.inc(57): OMV\Rpc\Rpc::call('UserMgmt', 'authUser', Array, Array, 2, true)
    #1 [internal function]: OMVRpcServiceSession->login(Array, Array)
    #2 /usr/share/php/openmediavault/rpc/serviceabstract.inc(123): call_user_func_array(Array, Array)
    #3 /usr/share/php/openmediavault/rpc/rpc.inc(86): OMV\Rpc\ServiceAbstract->callMethod('login', Array, Array)
    #4 /usr/share/php/openmediavault/rpc/proxy/json.inc(97): OMV\Rpc\Rpc::call('Session', 'login', Array, Array, 3)
    #5 /var/www/openmediavault/rpc.php(45): OMV\Rpc\Proxy\Json->handle()
    #6 {main}

    Before I attempt anything, I thought i'd ask the guru's how you would approach fixing this.


    I decided to try out AVAHI on the OMV server. It didn't work for whatever reason, so i immediately uninstalled it figuring i'd rather place it on another box anyway. I wasn't really paying attention and blindly hit Y and proceeded to unistall most of OMV by mistake...


    Whoops.


    Thoughts on the best way to attack this? Fresh Install?


    I say it depends on what you are going to use them for.

    For instance - my VM's live on one mirror that i leave it all turned off as they will be in use constantly.

    My DVR storage lives on a mirror with APM set to Intermediate, as they get recorded to multiple times a day and it's annoying hitting play and having to wait for the disks to spin up.

    My bulk storage mirror I set to minimum power with spindown, and 5 minutes spindown as the data is rarely accessed.


    What you don't want to do is to set aggressive power management policies on disks that are accessed all the time. You don't want to have the things spinning up and down constantly and having to wait for them to spin up every time you want to access a file.

    Great thanks.


    I plan to attempt the upgrade to 5.x & I had read of someone else running into issues upgrading while running proxmox kernel (not sure why but still). Plus I want the RAM back from ZFS so planned to remove anything ZFS related before the upgrade then create the vols again on btrfs post upgrade.

    I'm planning on ditching ZFS and subsequently will be removing the Proxmox kernel. Stupidly, at some point, I clicked the "remove non-proxmox kernels" button in the omv-extras plugin.


    As I'm trying to avoid heartache, can anyone recommend the best way to restore the correct kernel for OMV 4?

    OK it only seems to affect folders on ZFS volumes. I created some on an USB key with EXT4 and they survive a reboot.


    I guess the change has implemented a timing issue (even though the .mount file contains the dependancies)


    Running the following commands restores the shares which I assume means that all the config is OK


    Code
    systemctl daemon-reload
    systemctl restart local-fs.target

    Thanks. I appreciate the response.


    I've actually un-installed all plugins (apart from ZFS) to remove the possibility that one of them was interfering by starting earlier than the mount. I've also uninstalled the one LInux service (HDHomeRun) that i've installed.
    Unfortunately the problem remains.


    I'm happy to use the actual path, but many of the plugins like VirtualBox only allow you to select the "Shared Folder" path so it needs to survive reboots.


    I don't actually know where to go to next.

    Further - is the way that OMV works by writing these as mount-points in fstab?
    (sorry if my terminology is all mixed up, i'm new to Linux)


    I can see the shares defined in the /etc/openmediavault/config.xml


    for example




    but I don't see corresponding entries in the etc/fstab

    I did see a similar thread but it never seemed to be resolved.


    Every time I reboot now, all the "Shared Folders" appear empty.


    This is a setup I've been fooling with for a few months, with the intention of putting it into production. It's been rebooted numerous times without issue, but as of today, after updating to 4.1.24-1.and restarting, the shared folders no longer seem to have any files in them on each boot. The files still exist but not when accessed via the sharedfolders path.


    I can't see them if I SSH in as root.
    None of the services can see them - e.g VirtualBox, HDHomeRun.
    I CAN see the files via Samba, but looking at the smb.conf file it appears that the SMB share points directly to the physical path, not the sharedfolders path.


    I have two ZFS mirrors for storage, i can see all the data and files if I go to the pool directly.


    I can "fix" the issue temporarily by adding a new shared folder and applying the config which I assume triggers a process to update the config files.


    Not sure where to start - where is the mapping between the Shared Folder and the physical path defined?

    I lost a drive the other day, it simply "disappeared". Obviously I tested it, changed drive slots etc to prove that it was the drive.


    Funnily enough, using Seatools, it would fail a short test but pass a long test every time. I'd started with long-tests and really only selected a short-test by mistake after several successful long tests. I now plan to schedule regular short tests.

    So i've decided after a hiatus to come back to the world of OMV for my home NAS/Server appliance. I used OMV back in the 1.x and 2.x days but moved to something else for a while. But now I've decided to rebuild and come back to the light.


    What i can't decide on however is exactly what set of solutions to use for my storage.


    Hardware:
    Core i5 8500
    32GB RAM
    LSI MegaRAID 9272-8i 512M cache 6Gbps SATA/SAS Raid *


    Samsung 970 EVO Plus NVMe SSD 500GB
    Intel 128GB SSD (unknown model)
    4x Seagate 4TB HDD


    *Not installed yet, bought on ebay for $50.


    The case is now full, i'm out of ports. My storage requirements are:


    Static Media - 6TB
    DVR - up to 2TB
    3-4 VM's around 90GB each. They don't do much, 1 is a DVR, one is a VDI desktop, the rest just lab play machines.


    Due to the amount of storage I required, the only way to achieve enough is to use "RAID 5" unless i throw out the 4x HDD and start again.


    I was originally considering using ZFS with RaidZ1, which seems to perform pretty damn well. However I was concerned about the random IO performance for the VM's.
    Would I be best using the 500GB SSD as a separate disk just for VM's, or should I use it as a cache for the raidz volume?


    Then I have the RAID card.. which would make things simpler but I'm still worried about IO performance with the VM's and I really don't want the drives spun up 24x7.


    I think i've talked myself into raidz with the SSD as a cache, but what would you do?


    P.S the performance of that EVO SSD compared to the intel SSD is like comparing an SSD to a HDD. It's crazy quick.

    I've built a new 1.0 OMV box and moved in a RAID5 setup from my old system. It was created on a very early version of OMV (0.2?) and migrated through to 0.5 and then now to 1.0.x.
    Along the way the permissions have got a little screwed in some places, so I basically want to "reset" the entire drive to base level permissions and ownership. can someone clarify for me exactly what permissions I should set to keep everything happy? For instance I can't do anything in the folders with extplorer currently.

    So i've tested the plugin and so far it seems pretty good with the exception of the following issue


    - When going to the Static Entries tab, click the + button to add a static entry. Select an existing DHCP client lease. It does not fill in any of the fields (i.e MAC, IP, Host name etc) so when you click OK it errors out. In the old version selecting an existing lease would fill out all those fields. The new "Lease" tab also does not allow you to select the MAC so you can't even copy and paste it either.


    Using FF 28 and Chrome 34.something


    Cheers & thanks so much for finally getting this plugin near ready.