Installing on Linux

This chapter demonstrates how to use KubeKey to provision a production-ready Kubernetes and Kube AI Hub cluster on Linux in different environments. You can also use KubeKey to easily scale out and in your cluster and set various storage classes based on your needs.

Most Popular Pages

Below you will find some of the most viewed and helpful pages in this chapter. It is highly recommended that you refer to them first.

Introduction

Overview

Explore the general content in this chapter, including installation preparation, installation tool and method, and storage configurations.

KubeKey

Understand what KubeKey is and how it works to help you create, scale and upgrade your Kubernetes cluster.

Multi-node Installation

Learn the general steps of installing Kube AI Hub and Kubernetes on a multi-node cluster.

Air-gapped Installation

Understand how to install Kube AI Hub and Kubernetes in the air-gapped environment.

Port Requirements

Understand the specific port requirements for different services in Kube AI Hub.

Kubernetes Cluster Configurations

Customize your Kubernetes settings in the configuration file for your cluster.

Time Synchronization

Unify timezone and configure time synchronization on all nodes before installing Kubernetes to avoid ServiceAccount token 401 errors, certificate verification failures, and other issues caused by clock drift.

GPU Driver Installation

Overview and guidance for installing GPU/NPU drivers and container runtimes on Linux nodes for Kube AI Hub clusters.

NVIDIA

Install the NVIDIA GPU driver, NVIDIA Container Toolkit, and container runtime configuration on Linux GPU nodes.

Huawei Ascend

Install Huawei Ascend NPU drivers, firmware, and Ascend Docker Runtime on Linux NPU nodes.

High Availability Configurations

Set Up an HA Cluster Using the Internal HAProxy of KubeKey

Learn how to create a highly available cluster using the internal HAProxy of KubeKey.

Set up an HA Cluster Using a Load Balancer

Learn how to create a highly available cluster using a load balancer.

Set up an HA Cluster Using Keepalived and HAproxy

Learn how to create a highly available cluster using Keepalived and HAproxy.

Persistent Storage Configurations

Understand Persistent Storage Installation

Understand how to use KubeKey to install different storage systems.

Install QingCloud CSI

Use KubeKey to create a cluster with QingCloud CSI providing storage services.

Install NFS Client

Use KubeKey to create a cluster with NFS Client providing storage services.

Install GlusterFS

Use KubeKey to create a cluster with GlusterFS providing storage services.

Install Ceph

Use KubeKey to create a cluster with Ceph providing storage services.

Installing on Public Cloud

Deploy Kube AI Hub on Azure VM Instances

Learn how to create a high-availability cluster on Azure virtual machines.

Deploy Kube AI Hub on QingCloud Instances

Learn how to create a high-availability cluster on QingCloud platform.

Install and Use Kube AI Hub on Hostinger VPS

Learn how to use the Kube AI Hub VPS template.

Installing in On-premises Environments

Deploy Kube AI Hub on VMware vSphere

Learn how to create a high-availability cluster on VMware vSphere.

Deploy Kube AI Hub on Bare Metal

Learn how to create a multi-node cluster with one master on bare metal.

Deploy K3s and Kube AI Hub

Learn how to use KubeKey to install K3s and Kube AI Hub.

Installing Kube AI Hub on Minikube

Install Kube AI Hub on an existing Minikube cluster with a minimal installation package.

Add and Delete Nodes

Add New Nodes

Add more nodes to scale out your cluster.

Delete Nodes

Cordon a node and even delete a node to scale in your cluster.

Add Edge Nodes

Add edge nodes to your cluster.

Uninstall Kube AI Hub and Kubernetes

Remove Kube AI Hub and Kubernetes from your machines.