Metrics and Monitoring in CircleCI Server 4.x

Introduction

This article describes the options available for metrics and monitoring in CircleCI Server 4.x, and how that has evolved across versions.

Server 4.x through 4.9: 

Reference monitoring stack (Telegraf, Prometheus, Grafana)

From Server 4.0 through 4.9, CircleCI does not include a supported, built-in metrics solution. However, a reference monitoring stack is available as a separate Helm chart that provides a starting point for observability using Grafana, Prometheus, and Telegraf, along with pre-built dashboards for key service level indicators.

This reference stack is not an officially supported product. It is provided as a community reference implementation. CircleCI does not guarantee its compatibility across all versions or provide support for customizations.

You can find the reference monitoring stack and setup instructions here: Monitoring stack reference Helm chart.

Server 4.10 and later: OpenTelemetry collector

As of Server 4.10.0, Telegraf has been removed and replaced with an OpenTelemetry collector. If you are upgrading from 4.9.x or earlier and have customized your Telegraf configuration or built integrations around it, you will need to review and migrate those configurations as part of your upgrade.

Refer to the Server 4.10 release notes and the monitoring stack reference documentation for guidance on the new collector setup.

Additional Resources

• Monitoring stack reference Helm chart
• Server 4.10.0 release notes