Ready-to-use Podman Quadlet service definitions for systemd — a rootless, daemonless, compose-free way to run self-hosted services.

Each service is a plain .container file that systemd picks up natively via the Quadlet generator. No docker-compose, no podman-compose, no YAML — just INI files and systemctl.

.
├── configs/          Service configuration files (Caddyfile, YAML, TOML, JSON)
├── accessories/      Infrastructure and utility services
├── bookmarks/        Bookmark and read-later managers
├── git/              Git hosting
├── monitoring/       Monitoring, alerting, and observability stack
├── networks/         Shared Podman network definitions
├── notes/            Note-taking and knowledge base services
├── volumes/          Shared Podman volume definitions
├── rss/              RSS/Atom feed readers
├── vault/            Password managers
├── vpn/              VPN and proxy services
└── wiki/             Wiki engines

• Native systemd integration — services start on boot, restart on failure, show up in journalctl
• Rootless by default — no daemon, no root, no attack surface
• Declarative — .container files describe the desired state, systemd handles the rest
• Auto-updates — podman auto-update pulls new images for containers with AutoUpdate=registry
• No extra tools — ships with Podman 4.4+, nothing else to install

# 1. Copy Quadlet files and shared network
cp -r bookmarks/linkding ~/.config/containers/systemd/linkding/
cp networks/caddy.network ~/.config/containers/systemd/networks/

# 2. Create data directory and environment file
mkdir -p ~/volumes/linkding/data
cp bookmarks/linkding/env.example ~/volumes/linkding/.env
nano ~/volumes/linkding/.env

# 3. Reload and start
export XDG_RUNTIME_DIR=/run/user/$(id -u)
systemctl --user daemon-reload
systemctl --user start linkding

The repository mirrors the target server structure. Three types of files go to three different locations:

• configs/ — service configuration files (Caddyfile, config.yaml, settings.yml, etc.). Edit example.org placeholders and copy to the server. These are safe to track in git.
• env.example — template for .env files containing secrets (passwords, tokens, API keys). Copy to ~/volumes/<service>/.env and fill in real values. Never commit .env files.
• volumes/ — persistent data directories on the server (databases, uploads, caches). Not part of this repo.

• Podman 4.4+ with Quadlet support
• systemd user session with lingering enabled (see User lingering below)
• Privileged ports allowed for rootless users (see Privileged ports below)

1. Copy .container and .network files to ~/.config/containers/systemd/<service>/
2. Copy shared networks: cp networks/caddy.network ~/.config/containers/systemd/networks/
3. Copy config files (if the service has them): cp -r configs/<service> ~/.config/containers/systemd/configs/
4. Create data directories: mkdir -p ~/volumes/<service>/data
5. Create .env from template and fill in real values:

cp <category>/<service>/env.example ~/volumes/<service>/.env
nano ~/volumes/<service>/.env

6. Reload and start:

export XDG_RUNTIME_DIR=/run/user/$(id -u)
systemctl --user daemon-reload
systemctl --user start <service>

export XDG_RUNTIME_DIR=/run/user/$(id -u)

systemctl --user daemon-reload              # reload unit files
systemctl --user start <service>            # start
systemctl --user stop <service>             # stop
systemctl --user restart <service>          # restart
systemctl --user status <service> --no-pager  # status
journalctl --user -u <service> -f           # follow logs

Important. These are real-world issues encountered in production. Read this section before deploying.

sudo loginctl enable-linger $USER

sudo sysctl -w net.ipv4.ip_unprivileged_port_start=80

echo "net.ipv4.ip_unprivileged_port_start=80" | sudo tee /etc/sysctl.d/podman-privileged-ports.conf
sudo sysctl --system

export XDG_RUNTIME_DIR=/run/user/$(id -u)

systemctl --user restart <service>

UserNS=keep-id

UserNS=keep-id:uid=1000,gid=1000

mkdir -p ~/volumes/<service>/data

systemctl --user stop <service1> <service2> ...
systemctl --user start <service1> <service2> ...

# caddy.network
[Network]
NetworkName=caddy

# any .container file
[Container]
Network=caddy.network

podman exec caddy caddy reload --config /etc/caddy/Caddyfile

A minimal .container file:

[Unit]
Description=My Service
After=network-online.target

[Container]
Image=docker.io/library/myimage:latest
PublishPort=127.0.0.1:8080:80
Volume=%h/volumes/myservice/data:/data
EnvironmentFile=%h/volumes/myservice/.env
Network=caddy.network
AutoUpdate=registry

[Service]
Restart=always

[Install]
WantedBy=default.target

1. Fork the repository
2. Add a new service directory with .container file(s) and env.example
3. Follow existing conventions:
   • Config files in configs/<service>/, data volumes in %h/volumes/<service>/
   • Secrets in EnvironmentFile= (pointing to %h/volumes/<service>/.env), not inline
   • App containers on caddy.network, DB/Redis on internal <service>.network only
   • AutoUpdate=registry for public images
4. Open a pull request

MIT