Skip to content

Latest commit

 

History

History
52 lines (32 loc) · 1.66 KB

README.md

File metadata and controls

52 lines (32 loc) · 1.66 KB

🚀 MonKube: Multi-Cluster Monitoring Simplified

monkube is a learning-focused project that enables you to set up a multi-cluster monitoring system using Thanos. It follows a pull-based approach, where a central component collects data from agents in different clusters on demand.

🏗️ Architecture :

Thanos-prometheus architecture The Prometheus instances in all the clusters are managed by Prometheus Operator for easier management and service discovery.

🛠️ Quick Start Guide

Prerequisites

  • KinD v

1. Management Cluster Setup 🖥️

Prepare your management cluster with Thanos:

# Grant execution permissions
sudo chmod +x ./create-management-cluster.sh

# Create management cluster
./create-management-cluster.sh [CLUSTER_NAME]

# Verify Thanos installation
kubectl get pods -n monitoring

2. Workload Cluster Deployment 🌟:

# Make installation script executable
sudo chmod +x ./install_new_cluster.sh

# Install and configure new workload cluster
./install_new_cluster.sh [CLUSTER_NAME] [MANAGEMENT_CLUSTER_NAME]

# Subscribe cluster to Thanos
subscribe_cluster.py [CLUSTER_NAME]

you can check if the cluster is added by checking logs of thanos querier in the management cluster or running :

  • kubectl port-forward svc/querier -n monitoring 9090:9090

NOTE: thanos might take some around 1~2 minutes to discover the new cluster

  • To remove a cluster cluster from thanos target list : unsubscribe_cluster.py [NAME]

This project is designed to help users understand multi-cluster monitoring with Thanos and Prometheus Operator while ensuring an efficient and scalable setup. 🚀