(5/8) Self-host your Media Center On Kubernetes with Plex, Sonarr, Radarr, Transmission and Jackett


This article is part of the series Build your very own self-hosting platform with Raspberry Pi and Kubernetes

  1. Introduction
  2. Install Raspbian Operating-System and prepare the system for Kubernetes
  3. Install and configure a Kubernetes cluster with k3s to self-host applications
  4. Deploy NextCloud on Kuberbetes: The self-hosted Dropbox
  5. Self-host your Media Center On Kubernetes with Plex, Sonarr, Radarr, Transmission and Jackett
  6. Self-host Pi-Hole on Kubernetes and block ads and trackers at the network level
  7. Self-host your password manager with Bitwarden
  8. Deploy Prometheus and Grafana to monitor a Kubernetes cluster



Introduction

In the next article of this series, we will learn how to install and configure a Media Center onto our Kubernetes platform to automate the media aggregation and management and play our Media files. The Media Center will be composed of the following components:



Namespace

We are going to isolate all the Kubernetes objects related to the Media Center into the namespace media.

To create a namespace, run the following command:

$ kubectl create namespace media



Persistence

The first step consists in setting up a volume to store our media files and data required to run each component. If you followed the previous articles to install and configure a self-hosting platform using RaspberryPi and Kubernetes, you remember we have on each worker a NFS client pointing to a SSD on /mnt/ssd.

1. Deploy the Persistent Volume (PV)

The Persistent Volume specifies the name, the size, the location and the access modes of the volume:

Create the following file and apply it to the k8 cluster.

## media.persistentvolume.yml
---
apiVersion: v1
kind: PersistentVolume
metadata:
  name: "media-ssd"
  labels:
    type: "local"
spec:
  storageClassName: "manual"
  capacity:
    storage: "200Gi"
  accessModes:
    - ReadWriteMany
  hostPath:
    path: "/mnt/ssd/media"
---

$ kubectl apply -f media.persistentvolume.yml
persistentvolume/media-ssd created

You can verify the PV exists with the following command:

$ kubectl get pv

NAME            CAPACITY   ACCESS MODES   RECLAIM POLICY   STATUS      CLAIM   STORAGECLASS   REASON   AGE
media-ssd       200Gi      RWO            Retain           Available           manual                  34s


2. Create the Persistent Volume Claim (PVC)

The Persistent Volume Claim is used to map a Persistent Volume to a deployment or stateful set. Unlike the PV, the PVC belongs to a namespace.

Create the following file and apply it to the k8 cluster.

## media.persistentvolumeclaim.yml
---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  namespace: "media"
  name: "media-ssd"
spec:
  storageClassName: "manual"
  accessModes:
    - ReadWriteMany
  resources:
    requests:
      storage: "200Gi"
---

$ kubectl apply -f media.persistentvolumeclaim.yml
persistentvolumeclaim/media-ssd created

You can verify the PVC exists with the following command:

$ kubectl get pvc -n media

NAME            STATUS   VOLUME          CAPACITY   ACCESS MODES   STORAGECLASS   AGE
media-ssd       Bound    media-ssd       200Gi      RWO            manual         26s



Ingress

After the persistent volume, we are now going to deploy the ingress responsible of making accessible a service from outside the cluster by mapping an internal service:port to a host. To choose a host, we need to configure a DNS like we did for NextCloud "nextcloud." in the previous article. However, unlike NextCloud, the Media Center components have no reason to be exposed on the Internet, we can pick a host that will be resolved internally to our Nginx proxy (available at 192.168.0.240 : LoadBalancer IP). The simplest solution is to use nip.io which allows us to map an IP (in our case 192.168.0.240) to a hostname without touching /etc/hosts or configuring a DNS. Basically it resolves <anything>.<ip>.nip.io by <ip> without requiring anything else, Magic !

1. Create the file media.ingress.yaml

Create the following Ingress config file media.ingress.yaml to map the routes to each service we will deploy right after this step:

## media.ingress.yaml
---
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  namespace: "media"
  name: "media-ingress"
spec:
  rules:
  - host: "media.192.168.0.240.nip.io"
    http:
      paths:
        - backend:
            serviceName: "transmission-transmission-openvpn"
            servicePort: 80
          path: "/transmission"
        - backend:
            serviceName: "sonarr"
            servicePort: 80
          path: "/sonarr"
        - backend:
            serviceName: "jackett"
            servicePort: 80
          path: "/jackett"
        - backend:
            serviceName: "radarr"
            servicePort: 80
          path: "/radarr"
        - backend:
            serviceName: "plex-kube-plex"
            servicePort: 32400
          path: "/"
---


2. Deploy the ingress

Deploy the Ingress by applying the file media.ingress.yaml.

$ kubectl apply -f media.ingress.yaml
ingress.extensions/media-ingress created


3. Confirm the Ingress is correctly deployed

Try the URL http://media.192.168.0.240.nip.io from your browser and confirm it returns the error message 503 Service Temporarily Unavailable which is normal because we haven't deployed anything yet.



Heml Repository - add Bananaspliff

Someone already made a very good job at creating specific Helm Charts for the all the software we wish to install in this tutorial. Add the following repository to your Helm using the following command:

$ helm repo add bananaspliff https://bananaspliff.github.io/geek-charts
$ helm repo update




BitTorrent client - Transmission over VPN

The first bit of software to install is Transmission, an open-source BitTorent client offering an API, great for integration and automation. Because many Internet providers and Governments disproves BitTorent download, we are going to deploy Transmission alongside a VPN. The image haugene/transmission-openvpn includes Transmission and supports a very large range of VPN providers (see here) to obfuscate the traffic. I will be using NordVPN but change appropriately to your need.


1. Create a Kubernetes secret to store your VPN password

We first need to safely store our VPN Provider username and password into a Kubernetes secret. Run the command using your own VPN username and password:

$ kubectl create secret generic openvpn \
    --from-literal='username=<VPN_USERNAME>' \
    --from-literal='password=<VPN_PASSWORD>' \
    --namespace media


2. Write the Helm configuration

Next, we will configure the chart bananaspliff/transmission-openvpn. The default configuration can be seen by running the following command $ helm show values bananaspliff/transmission-openvpn.

Create the file media.transmission-openvpn.values.yml containing the following configuration.

## media.transmission-openvpn.values.yml
replicaCount: 1

image:
  repository: "haugene/transmission-openvpn"
  tag: "latest-armhf" # Suffixed by -armhf to pull the ARM image
  pullPolicy: "IfNotPresent"

env:
  - name: OPENVPN_PROVIDER
    value: "NORDVPN" # VPN provider. List of supported providers: https://haugene.github.io/docker-transmission-openvpn/supported-providers/
  - name: OPENVPN_USERNAME
    valueFrom: # Reference to the secret | openvpn.username
      secretKeyRef:
        name: "openvpn"
        key: "username"
  - name: OPENVPN_PASSWORD
    valueFrom: # Reference to the secret | openvpn.password
      secretKeyRef:
        name: "openvpn"
        key: "password"
  - name: NORDVPN_PROTOCOL
    value: "TCP"
  - name: NORDVPN_COUNTRY
    value: "CH" # Country where we want to download over VPN
  - name: NORDVPN_CATEGORY
    value: "P2P" # VPN Type
  - name: LOCAL_NETWORK
    value: "192.168.0.0/24"
  - name: TRANSMISSION_PEER_PORT
    value: "47444"
  - name: TRANSMISSION_DOWNLOAD_DIR
    value: "/downloads/transmission"
  - name: PUID
    value: "1000"
  - name: PGID
    value: "1000"

service:
  type: ClusterIP
  port: 80

volumes:
  - name: "media-ssd"
    persistentVolumeClaim:
      claimName: "media-ssd" # PersistentVolumeClaim created earlier
  - name: "dev-tun" # Needed for VPN
    hostPath:
      path: "/dev/net/tun"

volumeMounts:
  - name: "media-ssd"
    mountPath: "/data"
    subPath: "configs/transmission-data" # Path /mnt/ssd/media/configs/transmission-data where transmission writes the configuration
  - name: "media-ssd"
    mountPath: "/downloads/transmission"
    subPath: "downloads/transmission" # Path /mnt/ssd/media/downloads/transmission where transmission downloads Torrents
  - name: "dev-tun"
    mountPath: "/dev/net/tun" # Needed for VPN

securityContext:
  capabilities: # Needed for VPN
    add:
      - NET_ADMIN


2. Install the chart bananaspliff/transmission-openvpn

Execute the following command to install the chart bananaspliff/transmission-openvpn with the above configuration onto the namespace media.

$ helm install transmission bananaspliff/transmission-openvpn \
    --values media.transmission-openvpn.values.yml \
    --namespace media

After a couple of minutes, you should observe a pod named transmission-transmission-openvpn-xxx Running.

$ kubectl get pods -n media -l app=transmission-openvpn -o wide

NAME                                                 READY   STATUS    RESTARTS   AGE   IP           NODE           NOMINATED NODE   READINESS GATES
transmission-transmission-openvpn-8446dbf97c-rzw5l   1/1     Running   0          29m   10.42.1.26   kube-worker1   <none>           <none>


3. Access to Transmission Web console

Now Transmission and the Nginx Ingress routes are deployed, you should be able to access the Transmission Web console via http://media.192.168.0.240.nip.io/transmission.




Torrent Providers Aggregator- Jackett over VPN

Jackett is a Torrent Providers Aggregator which translates search queries from applications like Sonarr or Radarr into tracker-site-specific http queries, parses the html response, then sends results back to the requesting software. Because some Internet Providers might also block access to Torrent websites, I packaged a version of Jackett using a VPN connection (similar to transmission-over-vpn) accessible on Docker hub - gjeanmart/jackettvpn:latest-armhf.


1. Write the Helm configuration

Let's now configure the chart bananaspliff/jackett. The default configuration can be seen by running the following command $ helm show values bananaspliff/jackett.

Create the file media.jackett.values.yml containing the following configuration.

## media.jackett.values.yml
replicaCount: 1

image:
  repository: "gjeanmart/jackettvpn" # Special image to use Jackett over a VPN
  tag: "latest-armhf"
  pullPolicy: IfNotPresent

env:
  - name: VPN_ENABLED
    value: "yes" # Enable Jackett over VPN
  - name: VPN_USERNAME
    valueFrom:
      secretKeyRef: # Reference to the secret | openvpn.username
        name: "openvpn"
        key: "username"
  - name: VPN_PASSWORD
    valueFrom:
      secretKeyRef: # Reference to the secret | openvpn.password
        name: "openvpn"
        key: "password"
  - name: LAN_NETWORK
    value: "192.168.0.0/24"
  - name: CREATE_TUN_DEVICE
    value: "true" # Needed for VPN
  - name: PUID
    value: "1000"
  - name: PGID
    value: "1000"

service:
  type: ClusterIP
  port: 80

volumes:
  - name: "media-ssd"
    persistentVolumeClaim:
      claimName: "media-ssd" # PersistentVolumeClaim created earlier
  - name: "dev-tun"  # Needed for VPN
    hostPath:
      path: "/dev/net/tun"

volumeMounts:
  - name: "media-ssd"
    mountPath: "/config"
    subPath: "configs/jackett" # Path /mnt/ssd/media/configs/jackett where jackett writes the configuration
  - name: "media-ssd"
    mountPath: "/downloads"
    subPath: "downloads/jackett" # Path /mnt/ssd/media/downloads/jackett ???

securityContext:
  capabilities: # Needed for VPN
    add:
      - NET_ADMIN


*2. Configure VPN (only if you configured VPN_ENABLED=yes)

a. Create the following directory structure on your SSD

$ mkdir -p /mnt/ssd/media/configs/jackett/openvpn/

b. Copy one OpenVPN file (usually provided by your VPN provider) into the folder /mnt/ssd/media/configs/jackett/openvpn/

c. Create a file credentials.conf into the folder /mnt/ssd/media/configs/jackett/openvpn/ composed of two line (first one: username and second one password)

<VPN_USERNAME>
<VPN_PASSWORD>

3. Pre-configure Jackett

a. Create the following directory structure on your SSD

$ mkdir -p /mnt/ssd/media/configs/jackett/Jackett/

b. Create the file ServerConfig.json into the folder /mnt/ssd/media/configs/jackett/Jackett/ with the following content:

{
  "BasePathOverride": "/jackett"
}


4. Install the chart bananaspliff/jackett

Execute the following command to install the chart bananaspliff/jackett with the above configuration onto the namespace media.

$ helm install jackett bananaspliff/jackett \
    --values media.jackett.values.yml \
    --namespace media

After a couple of minutes, you should observe a pod named jackett-xxx Running.

$ kubectl get pods -n media -l app=jackett -o wide

NAME                      READY   STATUS    RESTARTS   AGE    IP           NODE           NOMINATED NODE   READINESS GATES
jackett-864697466-69xwt   1/1     Running   0          101s   10.42.1.29   kube-worker1   <none>           <none>


5. Access Jackett

Go to Jackett on http://media.192.168.0.240.nip.io/jackett and try to add one or more indexers.




TV Show Library Management - Sonarr

Sonarr is a TV Show library management tool that offers multiple features:


1. Write the Helm configuration

Let's now configure the chart bananaspliff/sonarr. The default configuration can be seen by running the following command $ helm show values bananaspliff/sonarr.

Create the file media.sonarr.values.yml containing the following configuration.

### media.sonarr.values.yml
replicaCount: 1

image:
  repository: linuxserver/sonarr
  tag: arm32v7-latest # ARM image
  pullPolicy: IfNotPresent

env:
  - name: PUID
    value: "1000"
  - name: PGID
    value: "1000"

service:
  type: ClusterIP
  port: 80

volumes:
  - name: media-ssd
    persistentVolumeClaim:
      claimName: "media-ssd" # PersistentVolumeClaim created earlier

volumeMounts:
  - name: media-ssd
    mountPath: "/config"
    subPath: "configs/sonarr" # Path /mnt/ssd/media/configs/sonarr where sonarr writes the configuration
  - name: media-ssd
    mountPath: "/downloads/transmission"
    subPath: "downloads/transmission" # Path /mnt/ssd/media/downloads/transmission where sonarr picks up downloaded episodes
  - name: media-ssd
    mountPath: "/tv"
    subPath: "library/tv" # Path /mnt/ssd/media/library/tv where sonarr moves and renames the episodes


2. Pre-configure Sonarr

a. Create the following directory structure on your SSD

$ mkdir -p /mnt/ssd/media/configs/sonarr/

b. Create the file config.xml into the folder /mnt/ssd/media/configs/sonarr/ with the following content:

<Config>
  <UrlBase>/sonarr</UrlBase>
</Config>


3. Install the chart bananaspliff/sonarr

Execute the following command to install the chart bananaspliff/sonarr with the above configuration onto the namespace media.

$ helm install sonarr bananaspliff/sonarr \
    --values media.sonarr.values.yml \
    --namespace media

After a couple of minutes, you should observe a pod named sonarr-xxx Running.

$ kubectl get pods -n media -l app=sonarr -o wide

NAME                      READY   STATUS    RESTARTS   AGE     IP           NODE           NOMINATED NODE   READINESS GATES
sonarr-574c5f85d7-n9jc6   1/1     Running   0          3m13s   10.42.1.30   kube-worker1   <none>           <none>


4. Access Sonarr

Go to Sonarr on http://media.192.168.0.240.nip.io/sonarr and start setting up the library automation. Refer to the wiki for more details.




Movie Library Management - Radarr

Radarr is a Movie library management tool that offers multiple features:


1. Write the Helm configuration

Let's now configure the chart bananaspliff/radarr. The default configuration can be seen by running the following command $ helm show values bananaspliff/radarr.

Create the file media.radarr.values.yml containing the following configuration.

## media.radarr.values.yml
replicaCount: 1

image:
  repository: "linuxserver/radarr"
  tag: "arm32v7-latest" # ARM image
  pullPolicy: IfNotPresent

env:
  - name: PUID
    value: "1000"
  - name: PGID
    value: "1000"

service:
  type: ClusterIP
  port: 80

volumes:
  - name: "media-ssd"
    persistentVolumeClaim:
      claimName:  "media-ssd" # PersistentVolumeClaim created earlier

volumeMounts:
  - name: "media-ssd"
    mountPath: "/config"
    subPath: "configs/radarr" # Path /mnt/ssd/media/configs/radarr where radarr writes the configuration
  - name: "media-ssd"
    mountPath: "/downloads/transmission"
    subPath: "downloads/transmission" # Path /mnt/ssd/media/downloads/transmission where radarr picks up downloaded movies
  - name: media-ssd
    mountPath: "/movies"
    subPath: "library/movies" # Path /mnt/ssd/media/library/movies where radarr moves and renames the movies


2. Pre-configure Radarr

a. Create the following directory structure on your SSD

$ mkdir -p /mnt/ssd/media/configs/radarr/

b. Create the file config.xml into the folder /mnt/ssd/media/configs/radarr/ with the following content:

<Config>
  <UrlBase>/radarr</UrlBase>
</Config>


3. Install the chart bananaspliff/radarr

Execute the following command to install the chart bananaspliff/radarr with the above configuration onto the namespace media.

$ helm install radarr bananaspliff/radarr \
    --values media.radarr.values.yml \
    --namespace media

After a couple of minutes, you should observe a pod named radarr-xxx Running.

$  kubectl get pods -n media -l app=radarr -o wide

NAME                      READY   STATUS    RESTARTS   AGE   IP          NODE           NOMINATED NODE   READINESS GATES
radarr-7846697889-jhqbz   1/1     Running   0          11m   10.42.2.2   kube-worker2   <none>           <none>


4. Access Radarr

Go to Radarr on http://media.192.168.0.240.nip.io/radarr and start setting up the library automation. Refer to the wiki for more details.




Media Server - Plex

Plex Media Server is a software to serve and stream your personal Media library (movies, TV show and music). It fetches the Media resources and builds up a catalogue accessible to any compatible players (Desktop/Mobiles) and transcodes the stream to the player.


In this section, we are going to deploy Plex Media Server (PMS) on Kubernetes using the Helm chart kube-plex.

1. Clone the charts

This Helm chart is not available via an online repository like jetstack or bananaspliff. We need to download the chart locally. Clone the following repository using git.

$ git clone https://github.com/munnerz/kube-plex.git


2. Get a claim token

Obtain a Plex Claim Token by visiting plex.tv/claim. You need to create an account if you haven't already one yet.

This will be used to bind your new PMS instance to your own user account automatically.


3. Create the Helm config file media.plex.values.yml

## media.plex.values.yml

claimToken: "<CLAIM_TOKEN>" # Replace `<CLAIM_TOKEN>` by the token obtained previously.

image:
  repository: linuxserver/plex
  tag: arm32v7-latest
  pullPolicy: IfNotPresent

kubePlex:
  enabled: false # kubePlex (transcoder job) is disabled because not available on ARM. The transcoding will be performed by the main Plex instance instead of a separate Job.

timezone: Europe/London

service:
  type: LoadBalancer # We will use a LoadBalancer to obtain a virtual IP that can be exposed to Plex Media via our router
  port: 32400 # Port to expose Plex

rbac:
  create: true

nodeSelector: {}

persistence:
  transcode:
    claimName: "media-ssd"
  data:
    claimName: "media-ssd"
  config:
    claimName: "media-ssd"

resources: {}
podAnnotations: {}
proxy:
  enable: false


4. Install Plex using Helm

Now install Plex with Helm specifying our config file media.plex.values.yml and the namespace media:

$ helm install plex kube-plex/charts/kube-plex/ \
  --values media.plex.values.yml \
  --namespace media

Wait until kube-plex (Plex Media Server) is up and running.

$ kubectl get pods -n media -l app=kube-plex -o wide

NAME                             READY   STATUS    RESTARTS   AGE   IP           NODE          NOMINATED NODE   READINESS GATES
plex-kube-plex-b76f8f478-4tn97   1/1     Running   0          55s   10.42.0.27   kube-master   <none>           <none>

You can find the Virtual IP attributed to Plex by MetalLB (in my case 192.168.0.241).

$ kubectl get services -n media -l app=kube-plex -o wide

NAME             TYPE           CLUSTER-IP    EXTERNAL-IP     PORT(S)                                      AGE   SELECTOR
plex-kube-plex   LoadBalancer   10.43.77.73   192.168.0.241   32400:30048/TCP,80:32383/TCP,443:31274/TCP   16m   app=kube-plex,release=plex


5. Router config (outside access only)

If you want to access remotely to your Media library, you will need to configure a port-forwarding to allow Plex to access your PMS.

Add a route to port-forward incoming requests on port 32400 to 192.168.0.241:32400 (Plex virtual IP assigned by MetalLB).


6. Setup Plex

Try now to access (from your network) to Plex Web Player on http://192.168.0.241:32400. You should see the setup wizard :

You can uncheck Allow me to access my media outside my home if you only want to use Plex within your home network.

Our Media will be accessible from the folder /data/.


Notes



Conclusion

In conclusion, you now have everything you need to automate and manage your Media and enjoy watching shows, movies or just listen some music !