Mastodon on CloudFlow

Learn how to run a Mastodon server at the edge for low latency and high availability. Perform the steps below using the Kubernetes dashboard or kubectl commands.

note

Before starting, create a new CloudFlow Project and then delete the default Deployment and ingress-upstream Service to prepare the project for your new deployment.

Prerequisites

• You need a Postgres database (try using PolyScale.ai to optimize access to a fixed-location Postgres database of your choice). See our tutorial on how PolyScale works with CloudFlow.
• Redis database (try a managed Redis, such as DigitalOcean's, or from the several listed here).
• Email delivery service or SMTP server.
• S3 bucket (try a distributed managed object store like Supabase, Backblaze, Synadia Jetstream Obj store, or Wasabi).

Deploy It

Create a CloudFlow deployment for the Mastodon server with a mastodon-deployment.yaml file, substituting the environment variables accordingly. This will direct CloudFlow to distribute the linuxserver/mastodon image.

mastodon-deployment.yaml

apiVersion: apps/v1
kind: Deployment
metadata:
  name: mastodon
  labels:
    app: mastodon
spec:
  replicas: 1
  selector:
    matchLabels:
      app: mastodon
  template:
    metadata:
      labels:
        app: mastodon
    spec:
      containers:
      - name: mastodon
        image: linuxserver/mastodon:4.0.2
        imagePullPolicy: Always
        lifecycle:
          postStart:
            exec:
              command:
              - "/bin/sh"
              - "-c"
              - >
                sleep 5;
                sed -i -e "s/\$scheme/'https'/" /config/nginx/site-confs/default.conf
        resources:
          requests:
            memory: "1000Mi"
            cpu: "1000m"
          limits:
            memory: "1000Mi"
            cpu: "1000m"
        ports:
        - containerPort: 80
        readinessProbe:
          httpGet:
            port: 80
            httpHeaders:
            - name: "Host"
              value: "mastodon.example.com"
          failureThreshold: 15
          initialDelaySeconds: 60
          periodSeconds: 20
        env:
        - name: PUID
          value: "1000"
        - name: PGID
          value: "1000"
        - name: TZ
          value: "America/New_York"
        - name: LOCAL_DOMAIN
          value: "mastodon.example.com"
        - name: REDIS_HOST
          value: "redis"
        - name: REDIS_PORT
          value: "6379"
        - name: DB_HOST
          value: "db"
        - name: DB_USER
          value: "mastodon"
        - name: DB_NAME
          value: "mastodon"
        - name: DB_PASS
          value: "mastodon"
        - name: DB_PORT
          value: "5432"
        - name: SECRET_KEY_BASE
          value: ""
        - name: OTP_SECRET
          value: ""
        - name: VAPID_PRIVATE_KEY
          value: ""
        - name: VAPID_PUBLIC_KEY
          value: ""
        - name: SMTP_SERVER
          value: "mail.example.com"
        - name: SMTP_PORT
          value: "25"
        - name: SMTP_LOGIN
          value: ""
        - name: SMTP_PASSWORD
          value: ""
        - name: SMTP_FROM_ADDRESS
          value: "notifications@example.com"
        - name: ES_ENABLED
          value: "false"
        - name: ES_HOST # optional
          value: "es"
        - name: ES_PORT # optional
          value: "9200"
        - name: ES_USER # optional
          value: "elastic"
        - name: ES_PASS # optional
          value: "elastic"
        - name: S3_ENABLED
          value: "false"
        - name: S3_BUCKET # optional
          value: ""
        - name: AWS_ACCESS_KEY_ID # optional
          value: ""
        - name: AWS_SECRET_ACCESS_KEY # optional
          value: ""
        - name: S3_ALIAS_HOST # optional
          value: ""
        - name: WEB_DOMAIN # optional
          value: "mastodon.example.com"

Apply this deployment resource to your Project with either the Kubernetes dashboard or kubectl apply -f mastodon-deployment.yaml.

tip

For a production Mastodon server, use Kubernetes Secrets as the values for private environment variables.

Expose It

Expose it on the internet, mapping the container's port 80.

ingress-upstream.yaml

apiVersion: v1
kind: Service
metadata:
    labels:
        app: ingress-upstream
    name: ingress-upstream
spec:
    ports:
    - name: 80-8080
      port: 80
      protocol: TCP
      targetPort: 80
    selector:
        app: mastodon
    sessionAffinity: None
    type: ClusterIP

Apply this service resource to your Project with either the Kubernetes dashboard or kubectl apply -f ingress-upstream.yaml.

See the pods running on CloudFlow's network with either the Kubernetes dashboard or kubectl get pods -o wide. The -o wide switch shows where your app is running according to the default AEE location optimization strategy. Your app will be optimally deployed according to traffic. In lieu of significant traffic, your deployment will be made to default locations.

Finally, follow the instructions that configure DNS and TLS.

See What You've Built

See the Mastodon server you've built by visiting the https://mastodon.example.com, substituting mastodon.example.com according to your DNS and HTTPS configuration.