Posts by gadric

  • Nvidia Quadro P620 as HW Transcoding in Plex

    This thread is created to add some additional options and way to address some issues beyond this thread:

    Thread
    Using Nvidia Quadro P600 GPU in OMV 6+ for Tdarr and Plex Transcoding
    I've sifted through at least 50 web posts on how to install the Nvidia drivers in order to get docker to recognize the driver and make Tdarr and Plex use the GPU to transcode my videos. I became more confused the more I read. Folks tend to make things more complicated than they need to be. This will probably work for other Nvidia cards as well.

    I hope this helps others.

    I've narrowed my install down to 4 major steps. I am not one to re-invent the wheel and even happier when I can mostly just cut…
    reverendspam


    1. I am currently on Linux kernel 6.1.0.21. This creats issue with the Debian Nvidia drivers. To fixt this, I install the official Nvidia Driver from the website. One syptome being that the command nvidia-smi does not work.


    1. Download the current version from nvidia.com. Current version can be found here: https://www.nvidia.com/Downloa…esults.aspx/224350/en-us/
    2. Upload the file to the OMV
    3. chmod +x ./NVIDIA-Linux-$(uname -m)-*.run to make the file executable
    4. Install dependencies:
      apt -y install linux-headers-$(uname -r) make gcc acpid dkms
    5. Start the setup by running ./NVIDIA-Linux-$(uname -m)-*.run
    6. Remember to say "YES" to in setup if asked if you want to disable "Nouveau"
    7. Then install the Nvidia Container Toolkit: https://docs.nvidia.com/datace…latest/install-guide.html


    2. After everything works, I realised that after a sleep and wake-up cycle GPU transcoding stops working. After some research I found out that this issue is known. To fix that you need to stop Plex before going to sleep and then to start it after wake up.


    a. Create /etc/systemd/system/wakeup-script.service with the following content (considering that your plex container's name is plex)

    Code
    [Unit]
Description=Run script on wakeup from sleep
After=suspend.target

[Service]
Type=oneshot
ExecStart="docker plex start"

[Install]
WantedBy=suspend.target


    b. Enable the service: Enable the service so that systemd runs it after every wake-up: systemctl enable wakeup-script.service

    c. Reload systemd: Reload the systemd daemon to recognize the new service: systemctl daemon-reload


    d. Create /etc/systemd/system/pre-sleep.service with the following content (considering that your plex container's name is plex):

    Code
    [Unit]
Description=Run a script before the system goes to sleep
Before=sleep.target

[Service]
Type=oneshot
ExecStart="docker stop plex"

[Install]
WantedBy=sleep.target

    e. Enable the service so that it runs before the system goes to sleep: sudo systemctl enable pre-sleep.service


    f. Reload systemd manager configuration: sudo systemctl daemon-reload

  • systemd-resolved missing

    Well, I have 2 OMV setups here. One OMV in bridge-mode, one in normal mode. In terms of base OS and OMV setup, they are the same (except network).


    On the normal one, I do not have these issues. On the bridge mode one, this is the 2nd time I had this issue. That made me suspicious.

  • systemd-resolved missing

    Problem solved. It seems to be related to the fact that I have a bridged Network set up in OMV (because of Home Assistant in KVM). Somehow system-resolved.service was removed and I had no Internet connection. Local network access was working.


    I downloaded system-resolved directly from Debian: https://packages.debian.org/bullseye-backports/amd64/systemd-resolved/download


    Then upload it to OMV and executed: dpkg -i [tt]systemd-resolved_252.4-1~bpo11+1_amd64.deb


    Reboot and everything worked again.

  • systemd-resolved missing

    Hi,


    I am currently facing a strange issue. I ran apt update in the cli and it always showed me this:



    Also ping google.com results in: ping: google.com: Name or service not known



    After a little bit of digging I found out that my systemd-resolved is missing:


    Code
    root@xxxxxx:~# systemctl status systemd-resolved
Unit systemd-resolved.service could not be found.


    how can I fix this? omv-firstaid did not help

  • [GUIDE] Use Signal Messenger as a notification service

    This guide was inspired by the [GUIDE] Use Telegram as notification service by subzero79. I used Telegram for a long time. But due to a more privacy and data protection approach of Signal, I switched to Signal Messenger.

    What you need

    • A free phone number
    • curl
    • Remote ssh access with sudo access
    • Docker


    Signal does not offer a bot service, so we need to have an extra number to act as the bot. There are a couple of service providers which offers free phone numbers. Should be easy to obtain an extra number.


    Once you have the extra number, connect to your OMV instance via ssh and then install docker:


    Code
    sudo apt-get update
sudo apt-get install docker-ce docker-ce-cli containerd.io


    bbernhard created a signal-cli wrapper which offers a rest-api: https://github.com/bbernhard/signal-cli-rest-api. This will be used to send the notifications over Signal.


    To start the port using this command

    Code
    docker run \
  --name "/signal-cli-rest-api" \
  --volume "<config_path>:/home/.local/share/signal-cli" \
  --restart "unless-stopped" \
  --port "8080:8080" \
  "bbernhard/signal-cli-rest-api:latest"


    To register the number, follow the official guide: https://github.com/bbernhard/s…ob/master/doc/EXAMPLES.md


    Once this is done, we can create our script


    nano /usr/share/openmediavault/notification/sink.d/10signal


    Then paste this content into the file:


    Save the file


    Then the script needs to be made executable:


    chmod 700 /usr/share/openmediavault/notification/sink.d/10signal

  • AutoShutDown Plugin (Version 5.1.7) is inactive when it wakes up from sleep

    Quote from nrand

    Can you try moving: /usr/lib/systemd/system-sleep/autoshutdown-restart to /lib/systemd/system-sleep/autoshutdown-restart as in the 5.1.7 update we move to systemd start-up what than pm-utils. if this work you should see the script (above autoshutdown-restart) write a log entry to /var/log/autoshutdown.log on the restart if this is not logged the the scripts if still falling for some reason. Also can you verify you are using 'systemctl suspend' command.


    Please leave feedback if the above works and raze bug in: https://github.com/OpenMediaVa…enmediavault-autoshutdown and i jump on a fix ASAP


    I got this working, but not with this solution.


    In /usr/lib/systemd/system-sleep/ and in /lib/systemd/system-sleep/ the same file existed.


    What I did: chmod +x /lib/systemd/system-sleep/autoshutdown-restart not the /usr/... .


    Now it seems to work again.

  • AutoShutDown Plugin (Version 5.1.7) is inactive when it wakes up from sleep

    Hi,


    I have a problem with the most recent version of autoshutdown plugin. If worked perfect since a couple of years. With the most recent version (5.1.7) it behaves strange. When the system wakes up from sleep mode, then the plugin is inactive. `systemctl status autoshutdown.service` Shows `inactive`. In this state, my system wont get back to sleep again. I have to activate is manually.


    Any ideas?

  • Custom Notificaiton with Signal Messenger does not work with sink

    Hi,


    I have been using custom notification with telegram following this guide: [GUIDE] Use Telegram as notification service. This works very good also with OMV5.


    However, I am thinking to switch to Signal Messenger. I thought I could use a similar approach. I have duplicated the Telegram OMV sink.d file and inserted this code for signal:


    Bash
    #!/bin/bash

sender="+49xxxxxx"
recipiant="+49yyyyyyy"

echo -e "${OMV_NOTIFICATION_SUBJECT} on the ${OMV_NOTIFICATION_DATE} \n\n $(cat ${OMV_NOTIFICATION_MESSAGE_FILE})" |signal-cli -u $sender send $recipiant


    However, this script seems not to be triggered.


    I tested already manual message sending with signal which works very well. Also using this commands it works without any issues:


    Code
    export OMV_NOTIFICATION_SUBJECT=test
export OMV_NOTIFICATION_DATE=testdate
export OMV_NOTIFICATION_MESSAGE_FILE=/tmp/testfile 

/usr/share/openmediavault/notification/sink.d/./20signal


    Any idea why it does not work automatically with custom notification and sink scripts?

  • EFI Partition (/boot/efi) shows full, even it is not

    Hi,


    Since a couple of days I ran into a strange „problem“. In OMV WebGUI or using df in the console, it reports my /dev/sda1 mounted in /boot/efi as nearly full



    However, Checking with du, orgparted  reports something completed different:


    Code
    root@xxxxxxxx:~# du -sh /boot/efi/
5.2M    /boot/efi/


    I already purged old Kernels and also did apt-get autoremove and rebooted a couple of times. Nothing helped so far.


    Anyone has an idea?

  • Error when access Snapraid plugin

    Quote from Altecice

    3.7.6 appears to have fixed everything but the Diff script is still issuing the same error at line 490.



    Same error here.

  • mergerfs 2.26.2 on OMV4 does not work with Plex

    ah I see. Maybe the unisonFS plugin is the problem. Can you please post your fstab config ? I would like to check if something in my config is wrong.


    my config look is this:


    Code
    /srv/dev-disk-by-label-xxxx:/srv/dev-disk-by-label-xxxx:/srv/dev-disk-by-label-xxxx:/srv/dev-disk-by-label-xxxx:/srv/dev-disk-by-label-xxxx:/srv/dev-disk-by-label-xxxx /srv/yyyyyyyyyyyyyyyyyyyyyyyyyyy fuse.mergerfs defaults,allow_other,direct_io,use_ino,category.create=mfs,minfreespace=500G 0 0