Trouble Accessing Plex with Dockge Compose; Networking?

Apps and Virtualization
1

Hi Everyone,

I’ve finally upgraded to 25.04 and now I need to painfully recreate all the apps I was using under TrueCharts. I’m just trying to start with Plex for now, and no matter what I do, I cannot seem to connect to the web UI.

I’m using Dockge to manage the compose files. Here’s the Plex compose:

services:
  plex:
    image: lscr.io/linuxserver/plex:latest
    container_name: yottaplex
    environment:
      - PUID=568
      - PGID=568
      - TZ=CST6CDT
      - VERSION=docker
      - UMASK=0022
      - NVIDIA_VISIBLE_DEVICES=all
      - PLEX_ALLOWED_NETWORKS=172.16.0.0/12,10.0.0.0/8,192.168.0.0/16
    expose:
      - 32400
    ports:
      - 32400:32400
    volumes:
      - /mnt/vol1/container_states/yottaplex/config:/config
      - /mnt/vol1/tv:/mnt/tv
      - /mnt/vol1/movies:/mnt/movies
      - /mnt/ssd_scratch/incomding_media:/mnt/incoming_media
      - /mnt/vol1/movies_rifftrax:/mnt/movies_rifftrax
      - /mnt/vol1/audio_books:/mnt/audio_books
      - /mnt/vol1/music:/mnt/music
      - /mnt/vol1/container_states/yottaplex/Chromecast.xml:/usr/lib/plexmediaserver/Resources/Profiles/Chromecast.xml
      - type: tmpfs
        target: /transcode
    devices:
      - /dev/dri:/dev/dri
    restart: unless-stopped
    runtime: nvidia
networks:
  ix-dockge_default:
    external: true

I have tried with, and without, the following defined:

  • networks
  • expose

I am under the impression that bridge networking is enabled by default. But how do I tell it to bridge to a particular interface? My server is using 172.16.42.25/24 on interface br0. I try to connect to http://172.16.42.25:32400/web and receive the browser error message that the address is unreachable.

I am currently trying to access from a different network, 172.16.44.0/24, that is routed to the other network. This is working to all hosts on the 42 network and nothing has changed other than upgrading TrueNAS.

I am able to access Dockge installed from the apps repo on 172.16.42.25 just fine.

Any hints?

2

is for intel gpu passthrough
If you’re using nvidia you have to use

deploy:
         resources:
            reservations:
              devices:
                - driver: nvidia
                  count: 1
                  capabilities: [gpu]
3

Thank you for the reply. I will work on the transcoding config later. The container is launching and Plex is running. I just cannot access it. I have applied your suggestions but it didn’t make a difference accessing the web UI of course.

4

is that the ip of your truenas or the internal ip of the docker container?
because you should access plex on ipoftruenas:32400/web

5

Yes, that’s the IP of the TrueNAS server.

6

that may be a problem because it could collide with the internal docker networks…

7

No, I changed the docker network to 10.32.0.0/12. I can access Dockge just fine. I tried the catalog version of Plex and cannot access its web UI either. :-/ I am wondering if the allowed networks setting is not working, since I’m connecting from a different network.

8

remove the allowed networks and retry, you can always add them in the plex webgui

9

No luck. I tried removing it entirely, and also just allowing the 172.16.0.0/12 network. Typically I need to at least allow 172.16.0.0/12 since I am on a different subnet than the server.

10

found a typo in that line: incomding_media

11

OK i played around with a stripped down version (because i obviously do not have your mountpaths) and got it to deploy via custom yaml directly in truenas.
I tried modifying your yaml.
Please try it:

version: "3.8"

services:
  plex:
    image: lscr.io/linuxserver/plex:latest
    container_name: yottaplex
    environment:
      PUID: 568
      PGID: 568
      TZ: CST6CDT
      VERSION: docker
      UMASK: 0022
      NVIDIA_VISIBLE_DEVICES: all
      PLEX_ALLOWED_NETWORKS: 172.16.0.0/12,10.0.0.0/8,192.168.0.0/16
    ports:
      - "32400:32400"
    volumes:
      - /mnt/vol1/container_states/yottaplex/config:/config
      - /mnt/vol1/tv:/mnt/tv
      - /mnt/vol1/movies:/mnt/movies
      - /mnt/ssd_scratch/incoming_media:/mnt/incoming_media
      - /mnt/vol1/movies_rifftrax:/mnt/movies_rifftrax
      - /mnt/vol1/audio_books:/mnt/audio_books
      - /mnt/vol1/music:/mnt/music
      - /mnt/vol1/container_states/yottaplex/Chromecast.xml:/usr/lib/plexmediaserver/Resources/Profiles/Chromecast.xml
      - type: tmpfs
        target: /transcode
    restart: unless-stopped
    runtime: nvidia
    networks:
      - plex-default
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              capabilities: [gpu]
12

Good catch, but that wouldn’t have prevented connectivity. I fixed it, but no difference.

I rebooted the server just to make sure it wasn’t some weird bridging issue that I couldn’t see, but nothing changed. I guess for now I’m going to have to revert back to 24.04 for now. I will be back at that location tomorrow and can try from the same network subnet.

13

The problem was simply trying to access from a different subnet. Thanks, Plex.

I realize now I could’ve just forwarded a port over ssh to the host, but of well. It’s up and running now.

Thanks for your help!