The `Dockerfile` describes how the container image is made, and the `entrypoint.sh` is executed after the container is started. <br>
In this example, [WireGuard](https://www.wireguard.com/) is integrated into the container itself, so it should be a run-and-go(/out-of-the-box) experience.<br>
From there either use the environment variables described below as parameters or use the Docker Compose file: `compose.yaml`.<br>
Be careful, the default generated WireGuard configuration file uses port 51820/udp. So make sure to use this port if you want to use it out of the box.<br>
Otherwise edit the configuration file in WGDashboard under `Configuration Settings` -> `Edit Raw Configuration File`.
> Otherwise you need to enter the container and edit: `/etc/wireguard/wg0.conf`.
To run the container, you can either pull the image from the Github Container Registry (ghcr.io), Docker Hub (docker.io) or build it yourself. The image is available at:
| `tz` | Timezone | `Europe/Amsterdam` | `America/New_York` | Sets the container's timezone. Useful for accurate logs and scheduling. |
| `global_dns` | IPv4 and IPv6 addresses | `9.9.9.9` | `8.8.8.8`, `1.1.1.1` | Default DNS for WireGuard clients. |
| `public_ip` | Public IP address | Retrieved automatically | `253.162.134.73` | Used to generate accurate client configs. Needed if container is NAT’d. |
| `wgd_port` | Any port that is allowed for the process | `10086` | `443` | This port is used to set the WGDashboard web port. |
| `username` | Any non‐empty string | `-` | `admin` | Username for the WGDashboard web interface account. |
| `password` | Any non‐empty string | `-` | `s3cr3tP@ss` | Password for the WGDashboard web interface account (stored hashed). |
| `enable_totp` | `true`, `false` | `true` | `false` | Enable TOTP‐based two‐factor authentication for the account. |
You can develop against WGDashboard locally by mounting the `src/` directory into the container. This lets you edit Python and frontend code on your host and see changes reflected immediately (with a service restart for Python).
Create a `docker/compose-local.yaml` alongside the existing `compose.yaml`:
```yaml
services:
wgdashboard:
image:ghcr.io/wgdashboard/wgdashboard:latest
restart:unless-stopped
container_name:wgdashboard
ports:
- 10086:10086/tcp
- 51820:51820/udp
volumes:
- aconf:/etc/amnezia/amneziawg
- conf:/etc/wireguard
- data:/data
# Mount local src for live editing
- ../src:/opt/wgdashboard/src
# Keep venv in a named volume so it isn't overwritten by the mount
- venv:/opt/wgdashboard/src/venv
cap_add:
- NET_ADMIN
volumes:
aconf:
conf:
data:
venv:
```
The key additions compared to the production compose file:
-`../src:/opt/wgdashboard/src` — mounts your local `src/` directory into the container so code changes are reflected without rebuilding the image.
-`venv:/opt/wgdashboard/src/venv` — keeps the Python virtual environment in a named Docker volume. Without this, the host mount would overwrite the venv created during image build.
To start the development container:
```bash
cd docker
docker compose -f compose-local.yaml up -d
```
After editing Python files (e.g. `src/dashboard.py`), restart the container to pick up changes:
```bash
docker restart wgdashboard
```
For frontend changes, install dependencies and rebuild the Vue app on your host:
```bash
cd src/static/app
npm install
npm run build
```
Then restart the container so it serves the updated dist files:
