Speedtest Metrics Exporter is a Prometheus exporter designed to collect and expose network performance metrics using speedtest.net servers. Its unique feature is the ability to test multiple servers in a single iteration, providing comprehensive network performance data across different geographical locations and service providers.

• Multi-Server Testing: Test multiple speedtest servers in a single iteration by specifying server IDs, enabling comprehensive network performance monitoring across different locations
• Automatic Server Selection: If no server IDs are specified, the exporter automatically selects the closest server for testing
• Comprehensive Metrics: Collects detailed metrics including download speed, upload speed, latency, jitter, and test duration
• Server Information: Provides rich server metadata including server name, sponsor/provider, and distance
• Configurable Update Intervals: Customize how frequently speed tests are performed
• Optional BasicAuth Protection: Secure your metrics endpoint with basic authentication
• Instance Labeling: Support for instance labels to distinguish between multiple exporter instances
• Memory Optimization: Implements garbage collection after each test to maintain optimal performance

The exporter provides the following metrics:

Each metric includes the following labels:

• server_id: Speedtest server identifier
• server_name: Name of the speedtest server
• server_sponsor: Server sponsor/provider
• server_distance: Distance to server in kilometers
• instance: Custom instance label (optional)

The exporter can be configured using environment variables or command-line arguments:

See how to find Speedtest.net server IDs for instructions on locating specific server IDs.

# With specific server IDs
speedtest-exporter --server-ids=1234,5678 --update-interval=60 --metrics-port=9090

# With basic auth protection
speedtest-exporter --metrics-protected --metrics-username=custom_user --metrics-password=custom_pass

# With default server selection (closest server will be chosen automatically)
speedtest-exporter

docker run -d \
  -e SERVER_IDS=1234,5678,9012 \
  -e UPDATE_INTERVAL=60 \
  -p 9090:9090 \
  kutovoys/speedtest-exporter

services:
  speedtest-exporter:
    image: kutovoys/speedtest-exporter
    environment:
      - SERVER_IDS=1234,5678,9012
      - UPDATE_INTERVAL=60
      - METRICS_PROTECTED=true
      - METRICS_USERNAME=custom_user
      - METRICS_PASSWORD=custom_password
    ports:
      - "9090:9090"

Add the following to your prometheus.yml:

scrape_configs:
  - job_name: "speedtest"
    static_configs:
      - targets: ["localhost:9090"]
    scrape_interval: 60m

To test multiple servers, specify their IDs in the SERVER_IDS environment variable or --server-ids argument:

docker run -d \
  -e SERVER_IDS=1234,5678,9012 \
  -p 9090:9090 \
  kutovoys/speedtest-exporter

This will:

1. Test server ID 1234 first
2. After completion, test server ID 5678
3. Finally, test server ID 9012
4. Wait for the configured update interval before starting the next iteration

The exporter maintains separate metrics for each server, allowing you to:

• Compare performance across different providers
• Monitor network quality to different geographical locations
• Identify regional network issues
• Track performance trends per server

• /metrics - Prometheus metrics endpoint
• /health - Health check endpoint
• / - Basic information page

Contributions to Speedtest Metrics Exporter are warmly welcomed. Whether it's bug fixes, new features, or documentation improvements, your input helps make this project better. Here's a quick guide to contributing:

1. Fork & Branch: Fork this repository and create a branch for your work.
2. Implement Changes: Work on your feature or fix, keeping code clean and well-documented.
3. Test: Ensure your changes maintain or improve current functionality, adding tests for new features.
4. Commit & PR: Commit your changes with clear messages, then open a pull request detailing your work.
5. Feedback: Be prepared to engage with feedback and further refine your contribution.

Happy contributing! If you're new to this, GitHub's guide on Creating a pull request is an excellent resource.