115 lines
3.2 KiB
Markdown
115 lines
3.2 KiB
Markdown
<!-- <% if (false) { %> -->
|
|
#### 💬 How to setup?
|
|
<!-- <% } %> -->
|
|
|
|
### 0. Prepare your server
|
|
|
|
You'll need a server with a recent version of [Docker](https://www.docker.com/).
|
|
|
|
### 1. Create a GitHub personal token
|
|
|
|
From the `Developer settings` of your account settings, select `Personal access tokens` to create a new token.
|
|
|
|
No additional scopes are needed.
|
|
|
|
|
|
|
|
### 2. Configure your instance
|
|
|
|
Fetch [settings.example.json](/settings.example.json) which contains all supported option.
|
|
```shell
|
|
wget https://raw.githubusercontent.com/lowlighter/metrics/master/settings.example.json
|
|
```
|
|
|
|
Edit `settings.json` to configure your instance.
|
|
|
|
```javascript
|
|
{
|
|
//GitHub API token
|
|
"token":"GITHUB_TOKEN",
|
|
//Other options...
|
|
}
|
|
```
|
|
|
|
If you plan to make your web instance public, it is advised to restrict its access thanks to rate limiter and access list.
|
|
|
|
### 3. Start docker container
|
|
|
|
Metrics docker images are published on [GitHub Container Registry](https://github.com/lowlighter/metrics/pkgs/container/metrics).
|
|
|
|
Configure the following variables (or hardcode them in the command in the next block):
|
|
```shell
|
|
# Select an existing docker image tag
|
|
VERSION=latest
|
|
# Path to configured `settings.json`
|
|
SETTINGS=/path/to/settings.json
|
|
# Port used internally (use the same one than in `settings.json`)
|
|
SERVICE_PORT=3000
|
|
# Port to publish
|
|
PUBLISHED_PORT=80
|
|
```
|
|
|
|
And start the container using the following command:
|
|
```shell
|
|
docker run -d --workdir=/metrics --entrypoint="" -p=127.0.0.1:$PUBLISHED_PORT:$SERVICE_PORT --volume=$SETTINGS:/metrics/settings.json ghcr.io/lowlighter/metrics:$VERSION npm start
|
|
```
|
|
|
|
### 4. Embed link into your README.md
|
|
|
|
Edit your repository readme and add your metrics image from your server domain:
|
|
|
|
```markdown
|
|
|
|
```
|
|
|
|
### 5. (optional) Setup your instance as a service
|
|
|
|
To ensure that your instance will restart if it reboots or crashes, you should set it up as a service.
|
|
This is described below for Linux-like systems which support *systemd*.
|
|
|
|
Create a new service file `/etc/systemd/system/github_metrics.service` and paste the following after editing paths inside:
|
|
|
|
```ini
|
|
[Unit]
|
|
Description=Metrics
|
|
After=network-online.target
|
|
Wants=network-online.target
|
|
|
|
[Service]
|
|
Type=simple
|
|
ExecStart=(command to run as service)
|
|
|
|
[Install]
|
|
WantedBy=multi-user.target
|
|
```
|
|
|
|
Reload services, enable it, start it and check if it is up and running:
|
|
|
|
```shell
|
|
systemctl daemon-reload
|
|
systemctl enable github_metrics
|
|
systemctl start github_metrics
|
|
systemctl status github_metrics
|
|
```
|
|
|
|
### Alternative option: run an instance locally (intended for testing and development)
|
|
|
|
To run an instance without docker, you'll need to have [NodeJS](https://nodejs.org) (see used version in [Dockerfile](/Dockerfile#L1-L2)) installed.
|
|
|
|
Clone repository, install dependencies and copy configuration example to `settings.json`:
|
|
|
|
```shell
|
|
git clone https://github.com/lowlighter/metrics.git
|
|
cd metrics/
|
|
npm install --only=prod
|
|
cp settings.example.json settings.json
|
|
```
|
|
|
|
Once you've finished configuring metrics, start your instance using the following command:
|
|
|
|
```shell
|
|
npm start
|
|
```
|
|
|
|
You should now be able to access your server with provided port in `setting.json` from your browser.
|