A set up script for Pleroma in Podman
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
Paul Wilde cf4158466f Update 'README.md' 7 months ago
config first commit for pleroma via podman 7 months ago
.dockerignore first commit for pleroma via podman 7 months ago
Dockerfile first commit for pleroma via podman 7 months ago
LICENSE first commit for pleroma via podman 7 months ago
README.md Update 'README.md' 7 months ago
config.exs first commit for pleroma via podman 7 months ago
podman-run.sh re-write of readme 7 months ago

README.md

Pleroma

Pleroma is a federated social networking platform, compatible with GNU social and other OStatus implementations. It is free software licensed under the AGPLv3.

It actually consists of two components: a backend, named simply Pleroma, and a user-facing frontend, named Pleroma-FE.

Its main advantages are its lightness and speed.

Pleroma

Pleromians trying to understand the memes

Features

This Podman set up is adapted from angristan's docker-pleroma. I muddled my way through the set up to get a final working set up. There may be some missed steps in the below, but ultimately this is what lead me to getting things running nicely.

A quick further note. I have included the config :pleroma, configurable_from_database: true configuration and the pleroma_ctl config migrate_to_db procedure. This means a lot of the actual config.exs parts can be eliminated completely, but I've left them in for clarity.

As with Angristan's Docker-Pleroma: this is not a reusable (e.g. It can't be uploaded to the Docker Hub), because for now Pleroma needs to compile the configuration. 😢 Thus you will need to build the image yourself, but I explain how to do it below.

Build-time variables

  • PLEROMA_VER : Pleroma version (latest commit of the develop branch by default)
  • GID: group id (default: 911)
  • UID: user id (default: 911)

Usage

Installation

Create a folder for your Pleroma instance. Inside, you should have Dockerfile and podman-run.sh from this repo.

You should change the POSTGRES_PASSWORD variable in the podman-run.sh file.

Create the upload and config folder and give write permissions for the uploads:
The podman-run.sh script does this automatically.

mkdir uploads config
chown -R 911:911 uploads

Pleroma needs the citext PostgreSQL extension, here is how to add it:

The podman-run.sh script does this automatically when run with the db-setup argument.

./podman-run.sh db-setup

This creates a pod and the postgresql container then runs the below

"You don't need to do this bit"
podman exec -i pleroma-db psql -U pleroma -c "CREATE EXTENSION IF NOT EXISTS citext;"
"You don't need to do that bit"

Configure Pleroma. Copy the following to config/secret.exs:

use Mix.Config

config :pleroma, Pleroma.Web.Endpoint,
   http: [ ip: {0, 0, 0, 0}, ],
   url: [host: "pleroma.domain.tld", scheme: "https", port: 443],
   secret_key_base: "<use 'openssl rand -base64 48' to generate a key>"

config :pleroma, :instance,
  name: "Pleroma",
  email: "admin@email.tld",
  limit: 5000,
  registrations_open: true

config :pleroma, :media_proxy,
  enabled: false,
  redirect_on_failure: true,
  base_url: "https://cache.domain.tld"

# Configure your database
config :pleroma, Pleroma.Repo,
  adapter: Ecto.Adapters.Postgres,
  username: "pleroma",
  password: "pleroma",
  database: "pleroma",
  hostname: "postgres",
  pool_size: 10

You need to change at least:

  • host
  • secret_key_base
  • email

Make sure your PostgreSQL parameters are ok.

You can now build the image. 2 way of doing it: Again, the podman-run.sh script has runtime parameters to do this.

./podman-run.sh build-setup

This builds the container image and does the following: Sets up the database:

"You don't need to do this"
podman exec pleroma-web mix ecto.migrate

and also... migrates config to the DB

podman exec pleroma-web /pleroma/bin/pleroma_ctl config migrate_to_db
"You don't need to do that"

Get your web push keys and copy them to secret.exs: Again, the podman-run.sh script has runtime parameters to do this.

./podman-run.sh gen-keypair

Which sets up the containers again and runs the following:

"You don't need to do this"
podman exec pleroma-web mix web_push.gen.keypair
"You don't need to do that"

Put the output in your secret.exs (may not be necessary due to DB migration of config, but I've left it here for clarity again)

You will need to build the image again, to pick up your updated secret.exs file: Once again, podman-run.sh has a parameter for it:

./podman-run.sh final-build

You can now launch your instance:

podman pod start pleroma-pod

Check if everything went well with:

podman logs -f pleroma-web

You can now setup a HAProxy or Nginx reverse proxy in a container or on your host by using the example Nginx config.

Final Notes

As with anything, I've only tested this on my systems, and the process of going through getting things working may have meant some steps are missing from this guide. At some point I will test this process again to ensure the script works well, but if anyone has any queries about it then let me know.

Other Container images

Here are other Pleroma Container images that helped me build mine: