Overview

Although the Kubernetes Management Operator on managed clusters provides integrated monitoring and visibility capabilities, organizations may prefer to deploy and operate their own "custom monitoring" stack.

This document describes how customers can configure and use the Kube Prometheus Stack to store and query critical operational metrics from their Kubernetes clusters. It installs a collection of Kubernetes manifests to simplify and streamline the deployment and configuration of Prometheus, Alert Manager, and Grafana to operationalize your cluster's metrics.

Important

The managed Prometheus monitoring components are tuned and optimized to ensure that they will NOT collide with a customer's Prometheus Operator deployment on managed clusters.

What Will You Do

In this multi-part recipe, you will perform the following:

Part 1

• Create a custom cluster blueprint with the "ingress-nginx" and "kube-prometheus-stack" addons
• Create a Kubernetes secret
• Apply the newly created cluster blueprint to your EKS Cluster
• Update your DNS record

Part 2

• Access Grafana and visualize your cluster's metrics

Assumptions

• You have already provisioned or imported one or more EKS clusters using the controller.
• You have access to a cert and it's private key to create a Kubernetes secret which we will use to expose Grafana.
• You have access to your hosted zone to create the necessary DNS record for accessing grafana.