Prometheus Tunnel

Prometheus Tunnel is a reverse proxy server that forwards requests to a remote Prometheus exporter. This project includes functionality for metrics collection and health checks, making it easy to monitor and manage your Prometheus exporter.

Features
Quick Start
Installation
Configuration
Prometheus Metrics
Health Checks
Prometheus Alert Rules
Development
- Running Tests
- Static Analysis Tools
Contributing
License
Community and Support
Example Use Cases

Features

Reverse Proxy: Forwards incoming requests to a specified Prometheus exporter.
Metrics Collection: Collects and exposes metrics for request count, duration, and response statuses.
Health Checks: Provides endpoints for health and readiness checks.
Dockerized: Easily deployable as a Docker container.
Helm Chart: Package and deploy the application using Helm.

Quick Start

Install the Prometheus Tunnel using Helm:

helm repo add supporttools https://charts.support.tools
helm upgrade --install prometheus-tunnel-server01 supporttools/prometheus-tunnel \
    --namespace monitoring \
    --create-namespace \
    --set settings.serverIP=192.168.0.3 \
    --set settings.serverPort=9182 \
    --set settings.name=server01

Note, please replace the following:

prometheus-tunnel-server01 with the name of the Helm release.
server_name with the name of the remote Prometheus exporter (e.g., server01).
192.168.0.3 with the IP address of external Prometheus exporter.
9100 with the port of external Prometheus exporter.

Installation

Prerequisites

Go 1.22 or higher
Docker
Kubernetes (optional, for deployment)
Helm (optional, for deployment)

Building from Source

Clone the repository:

git clone https://github.com/supporttools/prometheus-tunnel.git
cd prometheus-tunnel

Build the binary:
```
go build -o prometheus-tunnel .
```
Run the application:
```
./prometheus-tunnel
```

Using Docker

Build the Docker image:

docker build -t supporttools/prometheus-tunnel:latest .

Run the Docker container:

docker run -p 8080:8080 -e SERVER_IP=your-prometheus-exporter-ip -e SERVER_PORT=your-prometheus-exporter-port supporttools/prometheus-tunnel:latest

Using Helm

Add the Helm repository:

helm add repo supporttools https://charts.support.tools

Deploy using Helm:

helm upgrade --install prometheus-tunnel supporttools/prometheus-tunnel \
    --namespace monitoring \
    --create-namespace \
    --values values.yaml

Configuration

Environment Variables

SERVER_IP: IP address of the Prometheus exporter.
SERVER_PORT: Port of the Prometheus exporter.
METRICS_PORT: Port for exposing Prometheus metrics (default: 9182).
DEBUG: Enable debug logging (default: false).

Prometheus Metrics

The following metrics are exposed:

proxy_total_requests: Total number of requests received.
proxy_request_duration_seconds: Histogram of request durations.
proxy_response_status_total: Count of responses by status code.

Health Checks

/healthz: Health check endpoint.
/readyz: Readiness check endpoint.
/version: Version information endpoint.

Prometheus Alert Rules

Here is a sample PrometheusRule for monitoring the Prometheus Tunnel:

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: prometheus-tunnel
  labels:
    prometheus: prometheus-tunnel
spec:
  groups:
  - name: prometheus-tunnel.rules
    rules:
    - alert: HighRequestRate
      expr: rate(proxy_total_requests[5m]) > 100
      for: 5m
      labels:
        severity: warning
      annotations:
        summary: "High Request Rate"
        description: "The request rate has exceeded 100 requests per minute."

    - alert: SlowRequestDuration
      expr: histogram_quantile(0.99, rate(proxy_request_duration_seconds_bucket[5m])) > 1
      for: 5m
      labels:
        severity: warning
      annotations:
        summary: "Slow Request Duration"
        description: "99th percentile request duration is greater than 1 second."

    - alert: HighErrorRate
      expr: rate(proxy_response_status_total{status=~"5.."}[5m]) > 10
      for: 5m
      labels:
        severity: critical
      annotations:
        summary: "High Error Rate"
        description: "The rate of 5xx errors has exceeded 10 errors per minute."

    - alert: HighLatency
      expr: rate(proxy_request_duration_seconds_sum[5m]) / rate(proxy_request_duration_seconds_count[5m]) > 0.5
      for: 5m
      labels:
        severity: critical
      annotations:
        summary: "High Request Latency"
        description: "The average request latency is greater than 0.5 seconds."

    - alert: HighRequestVolume
      expr: sum(rate(proxy_total_requests[5m])) by (job) > 1000
      for: 5m
      labels:
        severity: warning
      annotations:
        summary: "High Request Volume"
        description: "The total request volume has exceeded 1000 requests per minute."

Development

Running Tests

To run the tests locally:

go test -v ./...

Static Analysis Tools

To run static analysis tools:

golint ./...
staticcheck ./...
gosec ./...

Contributing

Contributions are welcome! Please fork the repository and create a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for details.

Community and Support

For support and community interaction, you can join our Slack channel or open an issue on GitHub.

Example Use Cases

Prometheus servers managed by Argocd or FluxCD where editing the scape_configs to add external exporters is difficult.
Air-gapped environments where the Prometheus exporter is not directly accessible from the cluster and a HTTP/Socks Proxy is required.
Monitoring a Prometheus exporter that is behind a firewall and only accessible via a VPN.

# README