Skip to main content

Prometheus

Prometheus is an open-source systems monitoring and alerting toolkit designed for reliability and scalability. Developed by SoundCloud in 2012 and now part of the Cloud Native Computing Foundation, Prometheus has become one of the most popular monitoring solutions in the industry. It is known for its powerful data model, flexible query language (PromQL), and the ability to collect time-series data, making it ideal for monitoring dynamic and highly complex environments.

Prometheus works by scraping metrics from instrumented targets, which expose their metrics via HTTP endpoints. It then stores these metrics in a time-series database, allowing users to query and visualize the data using built-in tools or third-party integrations like Grafana. Prometheus also supports alerting, enabling users to define rules that trigger notifications when certain conditions are met, helping to ensure timely responses to potential issues.

Stalwart Mail Server supports exporting metrics via pull to Prometheus, providing seamless integration with this powerful monitoring toolkit. By enabling the Prometheus exporter, Stalwart Mail Server makes its metrics available through a dedicated endpoint that Prometheus can scrape at regular intervals.

The integration with Prometheus allows Stalwart Mail Server users to leverage Prometheus' advanced monitoring and alerting capabilities. Users can create custom dashboards to visualize the collected metrics, define alerting rules to monitor specific conditions, and use PromQL to perform detailed queries and analyses. This flexibility and power enable administrators to maintain optimal server performance, quickly detect and resolve issues, and make informed decisions based on real-time data.

Scraping Metrics​

When the Prometheus exporter is enabled, Stalwart Mail Server exposes its metrics under the /metrics/prometheus scraping endpoint. This endpoint provides a comprehensive set of metrics that reflect various aspects of the server's performance and health, including resource usage, request handling, and error rates.

Configuring Prometheus to scrape metrics from Stalwart Mail Server involves adding a job definition to the Prometheus configuration file. This definition specifies the target endpoint (i.e., the /metrics/prometheus URL) and the scraping interval. Once configured, Prometheus will automatically query the endpoint at the defined intervals, collect the metrics, and store them in its time-series database.

Configuration​

The following options are available under metrics.prometheus for configuring the Prometheus metrics pull exporter:

  • enable: Enables or disables the Prometheus metrics exporter. When set to true, Stalwart Mail Server will expose metrics for scraping by Prometheus. The default value is false.
  • auth.username: The username for basic authentication when accessing the Prometheus metrics endpoint. This option is optional.
  • auth.secret: The password or secret token for basic authentication when accessing the Prometheus metrics endpoint. This option is optional.

Example​

Here is an example configuration snippet for setting up the Prometheus metrics pull exporter in Stalwart Mail Server:

[metrics.prometheus]
enable = true

[metrics.prometheus.auth]
username = "prometheus"
secret = "password123"