Observability

This directory contains scrape targets, RabbitMQ alerting rules, Alertmanager configuration, and RabbitMQ dashboards.

Quick Start

If you don't have Prometheus and Grafana installed, the quickest way to try out RabbitMQ observability is as follows:

Make sure

1. kubectl (version 1.18+) pointing to any running Kubernetes cluster
2. helm (version 3+) being installed

Optionally, if you want to receive Slack notifications for RabbitMQ alerts, you will need a Slack Webhook URL (see here how to create one).

# Optionally, to receive Slack notifications on RabbitMQ alerts, set the name of the Slack channel and the Slack Webhook URL:
export SLACK_CHANNEL='#my-channel'
export SLACK_API_URL='https://hooks.slack.com/services/paste/your/token'

./quickstart.sh

This will install Prometheus Operator, Prometheus, kube-state-metrics, Alertmanager, Grafana and will set up RabbitMQ scrape targets, RabbitMQ alerting rules, Slack notifications, and RabbitMQ Grafana dashboards. Note that the quickstart.sh script is not a production-ready setup. Refer to the official Prometheus and Grafana documentation on how to deploy a production-ready monitoring stack.

Learn more on RabbitMQ monitoring in:

• RabbitMQ Prometheus documentation
• Operator monitoring documentation
• TGIR S01E07: How to monitor RabbitMQ?
• Notify me when RabbitMQ has a problem