Beiträge von reverendspam

When I updated Plex, I now see a specific setting to chose my video card. So, I definitely know the NVIDIA Drivers are working.

Screenshot 2024-02-03 112929.jpg

I have a p600 which uses same driver installation. Make sure the Nvidia Docker Container Toolkit is installed and point Plex container to the Nvidia Driver.

    deploy:
      resources:
        reservations:
          devices:
          - driver: nvidia
            count: all
            capabilities: [gpu] 

I just posted a guide here 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 an paste code.

The instruction below worked for my setup. Your mileage may vary.

1) Install the Nvida drivers

2) Install the Nvidia Container Toolkit

3) Don't allow OMV to overwrite /etc/docker/daemon.json

4) Install Tdarr and Plex through OMV>Services>Compose>Files

1) - Check what version of Debian you are running and install the driver

lsb_release -d

You should get something like this:

root@omv6:~# lsb_release -d

Description: Debian GNU/Linux 11 (bullseye)

root@omv6:~#

- Follow the very easy instruction in the Debian wiki to install the driver. https://wiki.debian.org/Nvidia…#Debian_11_.22Bullseye.22

2) Install the Nvidia Container Toolkit following the very easy instructions on their website.

Nvidia Container Toolkit

Once the driver and toolkit is installed, you should a similar output to the one below.

Screenshot 2023-11-23 183656.jpg

3) NOTE: Upon OMV reboot and updates - the /etc/docker/daemon.json will be overwritten.

To stop this:

Go into your OMV GUI /Services/Compose/Settings

In the "Docker storage" Section remove /var/library/docker and then save.

You can then reboot the server to test that the daemon.json file is not being overwritten.

Screenshot 2024-02-03 105518.png

4) Install Tdarr through OMV>Services>Compose>Files

- Be careful about format/spacing. Compose is finicky and may not start if you have a heading in the wrong space.

The original example came from the Tdarr Compose section. I just had to tweak them for my OMV setup.

Example from my system. Remember change the environment variables and volume section parameters to your OMV system.

version: "3.4"
services:
  tdarr:
    container_name: tdarr
    image: ghcr.io/haveagitgat/tdarr:latest
    restart: unless-stopped
    network_mode: bridge
    ports:
     - 8265:8265 # webUI port
     - 8266:8266 # server port
    environment:
      - TZ=America/New_York
      - PUID=0
      - PGID=0
      - UMASK_SET=002
      - serverIP=192.168.1.5
      - serverPort=8266
      - webUIPort=8265
      - internalNode=true
      - inContainer=true
      - ffmpegVersion=6
      - nodeName=MyInternalNode
      - NVIDIA DRIVER CAPABILITIES=all
      - NVIDIA VISIBLE DEVICES=all
    volumes:
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/tdarr/server:/app/server
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/tdarr/configs:/app/configs
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/tdarr/logs:/app/logs
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/media2:/media
      - /srv/transcoder:/temp
    devices:
      - /dev/dri:/dev/dri
    deploy:
      resources:
        reservations:
          devices:
          - driver: nvidia
            count: all
            capabilities: [gpu]  
   
  tdarr-node:
    container_name: tdarr-node
    image: ghcr.io/haveagitgat/tdarr_node:latest
    restart: unless-stopped
    network_mode: service:tdarr
    environment:
      - TZ=America/New_York
      - PUID=0
      - PGID=0
      - UMASK_SET=002
      - nodeName=MyExternalNode
      - serverIP=192.168.1.5
      - serverPort=8266
      - inContainer=true
      - ffmpegVersion=6
      - NVIDIA DRIVER CAPABILITIES=all
      - NVIDIA VISIBLE DEVICES=all
    volumes:
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/tdarr/configs:/app/configs
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/tdarr/logs:/app/logs
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/media2:/media
      - /srv/transcoder:/temp
    devices:
      - /dev/dri:/dev/dri
    deploy:
      resources:
        reservations:
          devices:
          - driver: nvidia
            count: all
            capabilities: [gpu]

- Install PLEX through OMV>Services>Compose>Files

version: "2.1"
services:
  plex:
    image: ghcr.io/linuxserver/plex
    container_name: plex
    network_mode: host
    environment:
      - PUID=0
      - PGID=0
      - VERSION=docker
      - PLEX_CLAIM= #optional
      - NVIDIA_VISIBLE_DEVICES=all
      - NVIDIA_DRIVER_CAPABILITIES=compute,video,utility
    volumes:
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/compose/plex:/config
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/media2/videos/tvshows:/tv
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/media2/videos/movies:/movies
      - /srv/dev-disk-by-uuid-5c2eeef7-116c-4957-8428-cd205efd1bd2/media2/music/flac_clean:/music
      - /srv/transcoder/plex:/transcode
    devices:
      - /dev/dri:/dev/dri
    deploy:
      resources:
        reservations:
          devices:
          - driver: nvidia
            count: all
            capabilities: [gpu] 
    restart: unless-stopped

If you are the kind of person that can't be satisfied with just the basics, here is the explanation of the various gpu settings that you can use in your compose yaml file.

Me, I like to set it and forget it once it all works. My server has been sitting in a closet for 8 years and I just took it out to update the eight ( 8 year old) 4TB drives in a raid 6 array that have never failed me. The drives were getting a little slow an cranky. 4 out of the 8 drives were throwing a lot of SMART errors. I also neede more space. I ended up adding five 14TB drives in a raid 6 array. I hadn't updated OMV is a while. I figured I might as well update while I am building a new array. I did not realize OMV got rid of yacht and portainer...so here I am.

Once I linked my GPU to Tdarr in the compose file, my Tdarr CPU transcode speed went from the CPU 23fps to the Nvidia GPU cranking out 8000-13,000fps.

Screenshot 2023-11-23 125131.jpg

Once I linked my GPU to the Plex container it now shows hardware (hw) transcode when needed.

Screenshot 2023-11-23 135645.jpg

Tdarr Plugin GPU Transcode Setting

Make sure to use a Tdarr GPU Plugin that uses your GPU and set the switch to use "NVENC". Here is my Tdarr plugin stack that transcodes using the GPU.

Screenshot 2023-11-25 105440.jpg

I've seen so many installs of piwigo and I could not get any of them to really work properly on OMV5.

There are different ways to install this and you can upload your photos outside of the piwigo container, but I did not want to mess with symbolic links. I am totally ok with having piwigo and all my photos in one directory. This is just the quickest and easiest way to get piwigo up and running on OMV5.

- This tutorial assumes you have OMV5 installed and running properly.

- This tutorial assumes you have Portainer installed and running properly.

- This tutorial assumes you know how to use OMV5 and Portaineer.

1) Open portainer in your web browser.

- Click on Stacks --> Add Stack

- Give it a name such as piwigostack

- Copy and paste the code below and add it to the stack's Web Editor section. You will need to change the environment and volume parameters to suit your system. I have noted the main points you need to edit. Take out the comments before you deploy the stack.

version: 2.1
services:
  piwigo:
    image: linuxserver/piwigo
    container_name: piwigo
    environment:
      - PUID=1000   ##change to your user id. It cannot be the root user
      - PGID=100   ##change to your user group ID
      - TZ=America/New_York
    volumes:
      - /srv/dev-disk-by-label-myraid/media/piwigo:/config   ##this is the omv shared folders absolute path to where you have your media
    ports:
      - 8125:80
    restart: unless-stopped
  mariadb:
    image: linuxserver/mariadb
    container_name: piwidb
    environment:
      - PUID=0   ##change to your user id. I use root since I am on a closed system
      - PGID=0   ##change to your user group ID. Again I use root since I am on a closed system
      - MYSQL_ROOT_PASSWORD=whatever_you_want ##set a password
      - TZ=America/New_York
      - MYSQL_DATABASE=piwigo
      - MYSQL_USER=piwigo
      - MYSQL_PASSWORD=piwigo
    volumes:
      - /srv/dev-disk-by-label-myraid/appdata/piwidb:/config  ##point this to your omv5 shared folders absolute path
    ports:
      - 3306:3306
    restart: unless-stopped

##remeber to remove the comments before you deploy the stack.

- disable the "enable access control"

- Click "Deploy the stack"

You will now see 2 new containers named "piwidb" and "piwigo"

2) If everything went as planned open a web browser and point to your OMV5 ip at port 8125

for example, mine is : http://192.168.1.5:8125

You should have the piwigo instaltion screen in front of you.

3) Before you move forward you need to know the ip of your piwigo database.

- Go back to Portainer and under containers you should see the new "piwidb" container. Under the "quick actions" column open the exec_console for piwidb then connect.

- Once connected type in "hostname -I" without the quotation marks. This will give you your host database IP address.

- you see in my example below mine is 172.17.0.3

4) Go back to the piwigo installation install screen in your web browser and fill in the fields.

In the "Database Configuration" section:

- enter the host ip address of your database

- User is piwigo

- Password is piwigo

- database name is piwigo

- database tables prefix is piwigo_

In the "Administration Configuration" section:

- you enter whatever you want in this section

When you have all the fields filled in then click on "start installation"

If all goes well then you will get " Congratulations. Piwigo installation is completed"

That's it.

Note: I installed piwigo files to where my media files are housed because piwigo saves everything in one directory. The uploaded photos are saved in \www\gallery\upload.

I set my system up different ways at first until I settled on what I did below and my system smokes with very little CPU overhead. Best practice is whatever works best for you

SSD drives are cheap now. I bought a crucial 240 GB drive for $40. I installed OMV on it and expanded the partition with Gparted so I could run everything in a nice large partition drive.

I have been running this 24/7 with no crashes. People will tell you that this will basically burn out the ssd, but they are so cheap I do not care.

I purchased another ssd the same size and backed up my working system to it and set it aside for whenever it does crash. My system has been going strong for about 1 year now with no hiccups.

I was initially having trouble with plex not playing back my videos and locking up. I found out plex was, by default, writing the transcoded files to my hard drives instead of to my ssd drive thus having i/o conflicts. I ended up making a transcode partition on the ssd drive and directed plex and omv to write to it. I think emby is the same.

I can now play high def videos / audio with no issue. I even have one other person stream through the net off my system with no issues while I simultaneously stream through wifi at home.

I also had plex write it's database to my hard drives because it will eat up the ssd drive space if you have lots of media.

128gb ssd will be fine. I am only using about 40gb that includes omv, docker programs, and space for transcoding. This will increase during download activities.

Be aware that if you use one big drive for everything and download to it, remember to delete orphaned files leftover in your download folder from various programs that may have not automatically deleted or moved these orphaned files.

Good luck.

It could be any number of things. I have been running a nas at my house for 10 years now.

What I have found through trial and error and a lot of pulling my hair out;

1) I made the mistake of buying the cheapest sata cables and cursed my hard drives for being bad. I changed the cables out to better made ones and the drives worked flawlessly. Even new cables can be bad as it appears no one does quality control anymore to check products work out the door.

2) Drives, even new ones, can have bad sectors. Usually your manufacturer will have software that can be downloaded from their
website that will allow you to check the drives for bad sectors and 0 out any data that may be on them. This is a time consuming process.

3) Bad memory. I bought more expensive ECC memory that my mb manufacturer recommended. I made the mistake of throwing in memory I had laying around. It initially appeared to run fine, but it was causing data rot and corruption. If you do use non ecc memory or something you have laying around, run it through your bios memory checker if available and software made to run memory through its paces.

Joe

I could not get any of the solutions above to work. A post I found here worked for me and I updated sab to version 1.1.0

1) install the sab plugin as normal through gui

2) ssh into OMV and open the file /var/lib/dpkg/info/openmediavault-sabnzbd.postinst

nano  /var/lib/dpkg/info/openmediavault-sabnzbd.postinst

change the line that looks similar to

STABLE=`wget -q http://sabnzbd.org/download/ -O - | \grep sourceforge\.net | grep Linux | \sed 's/.*href=\"\(.*\)\".*/\1/' | sed 's/.*-\(.*\)-.*/\1/'`

to

STABLE="1.1.0"

ctrl o to write/save the file

then ctrl x to exit

finally force install

apt-get install -f