Monitoring #

Monitoring is essential for ensuring Forwarder’s health, performance, and resource utilization. This document provides a guide on how to collect Prometheus metrics, and setup Grafana.

Dashboard #

The Forwarder dashboard provides a high-level overview of health and performance. It consists of the following sections:

  • Errors - general information about the Forwarder health
  • HTTP - HTTP traffic metrics
  • Resources - memory, CPU, and golang runtime metrics
  • Info - version information and build details

Dashboard is available in Grafana dashboards registry. You can also use the source code version from here.


Import Dashboard from Grafana Registry #

If you are familiar with Prometheus and Grafana stack, you can import the dashboard with the following steps:

  1. Click Dashboards in the left-side menu.
  2. Click New and select Import in the dropdown menu.
  3. Perform one of the following steps:
    • Paste a Grafana dashboard ID 20100 and click Load.
    • Paste dashboard JSON text from here directly into the text area.

Step-by-Step Installation #

Forwarder exposes Prometheus metrics in the API server on the /metrics endpoint. The API server listens on localhost:10000 by default. This port will be used in the following steps.

Configure Prometheus #

  1. Install and start Prometheus following this guide.
  2. Configure Prometheus to scrape metrics from your Forwarder:
    • Open your Prometheus configuration file (usually prometheus.yml).
    • Add a new job to the scrape_configs section:
- job_name: 'forwarder'
  scrape_interval: 15s
    - targets: ['localhost:10000']

Configure Grafana #

  1. Install and start Grafana following this guide.
  2. Add Prometheus as a data source:
    • Go to Settings > Data Sources.
    • Click Add your first data source.
    • Choose Prometheus and configure the URL (e.g., http://localhost:9090).
  3. Import Dashboard from Grafana Registry:
    • Go to the Grafana Dashboard.
    • Click on the New or + icon select Import.
    • Type the following ID to import the dashboard: 20100.

Congratulations! You have successfully set up monitoring for your Forwarder using Prometheus and Grafana. Adjust the configuration parameters as needed based on your specific setup.

Automatic Local Setup #

We have provided a local monitoring setup that you can use to try out the monitoring features of the Forwarder. All you need is a Docker to run Prometheus, Grafana and Forwarder. See local/monitoring for more details.

© 2024 Sauce Labs Inc.