Kubernetes cluster deployment (v1.23.5)

1. Deployment Topology

1.1 Service configuration and topology

Mirror image ：CentOS-7-x86_64-Minimal-1908

CPU：2 nucleus

Memory ：2G

1.2 Change host name

// stay master Virtual machine , Set up master01 node

hostnamectl set-hostname master01

// Take effect immediately

bash

// see /etc/hostname by master01

cat /etc/hostname

// It will take effect permanently after restart

reboot

// In the same way node1 Virtual machine , Set up node01 node

hostnamectl set-hostname node01

// Take effect immediately

bash

// see /etc/hostname by node01

cat /etc/hostname

// It will take effect permanently after restart

reboot

node2 Such as node1 operation

2. install docker

// install yum Tools

yum install -y yum-utils

// Uninstall old version Docker

sudo yum remove docker docker-client docker-client-latest docker-common docker-latest docker-latest-logrotate docker-logrotate docker-engine

// Installation warehouse

sudo yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo

// Update configuration yum After source , Generate the cache with the following command , Improve the speed of searching and installing software

sudo yum makecache fast

// install Docker

sudo yum install docker-ce docker-ce-cli containerd.io -y

// start-up docker

sudo systemctl start docker

// Set boot up

sudo systemctl enable docker

sudo systemctl status docker

// New configuration file

cat  >> /etc/docker/daemon.json << EOF

{

 "exec-opts":["native.cgroupdriver=systemd"]

}

EOF

// restart docker

systemctl restart docker

systemctl status docker

3. Server configuration

3.1 Turn off firewall

systemctl stop firewalld

systemctl disable firewalld

3.2 take SELinux Set to permissive Pattern （ It's equivalent to disabling ）

setenforce 0

sed -i 's/^SELINUX=enforcing$/SELINUX=permissive/' /etc/selinux/config

3.3 allow iptables Check bridge flow

cat <<EOF | sudo tee /etc/modules-load.d/k8s.conf

br_netfilter

EOF

cat <<EOF | sudo tee /etc/sysctl.d/k8s.conf

net.bridge.bridge-nf-call-ip6tables = 1

net.bridge.bridge-nf-call-iptables = 1

EOF

sysctl --system

3.4 Shut down the system Swap

swapoff -a

// close swap Partition

sed -ri 's/.*swap.*/#&/' /etc/fstab

3.5 To configure /etc/hosts

// Customize master And node IP, Please modify... According to your personal situation

cat >> /etc/hosts << EOF

192.168.186.155 master01

192.168.186.156 node01

192.168.186.157 node02

EOF

4. install kubelet、kubeadm and kubectl

4.1 Configure software source

cat <<EOF > /etc/yum.repos.d/kubernetes.repo

[kubernetes]

name=Kubernetes

baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/

enabled=1

gpgcheck=1

repo_gpgcheck=0

gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg

EOF

// For modification repo_gpgcheck=0

4.2 install kubelet kubeadm kubectl

Install on all nodes kubelet、kubeadm and kubectl.

1. kubelet Running on the Cluster On all nodes , Responsible for starting Pod And the container .
2. kubeadm For initialization Cluster.
3. adopt kubectl Can be deployed and managed application , View resources , establish 、 Delete and update various components .

sudo yum update -y

//sudo yum install -y kubelet-1.19.4 kubeadm-1.19.4 kubectl-1.19.4

// According to github The release , Appoint

// Abreast of the times 1.24 Version is deprecated docker, Therefore, you need to specify 1.23

sudo yum install -y kubelet-1.23.5 kubeadm-1.23.5 kubectl-1.23.5

// Default latest version

//sudo yum install -y kubelet kubeadm kubectl

sudo systemctl enable --now kubelet

sudo systemctl start kubelet

//sudo systemctl status kubelet here kubelet Not ready yet , stay kubeadm init after master Nodes will ok, take node node join After adding kubelet It will be normal

4.3 Check tool installation

yum list installed | grep kubelet

yum list installed | grep kubeadm

yum list installed | grep kubectl

// View the cluster version results Kubernetes v1.23.5

kubelet --version

5. use kubeadm establish Cluster

5.1 initialization Master

1. --apiserver-advertise-address Indicate the use Master Which one? interface And Cluster Other node communications for . If Master There are many. interface, It is suggested to specify clearly , If you don't specify ,kubeadm Will automatically select the interface.
2. --pod-network-cidr Appoint Pod Scope of the network .Kubernetes Support multiple network solutions , And different network solutions --pod-network-cidr Have their own requirements , I'm going to set it to 10.244.0.0/16 Because we will use flannel Network plan , Must be set to this CIDR. In the later practice, we will switch to other network solutions , such as Canal.

//apiserver-advertise-address 192.168.186.155 by master node IP, According to the individual master IP modify

kubeadm init --apiserver-advertise-address 192.168.186.155 --image-repository registry.aliyuncs.com/google_containers --pod-network-cidr 10.244.0.0/16 --service-cidr 10.96.0.0/12

Add ：Kubeadm Initialization error ：[ERROR CRI]: container runtime is not running

terms of settlement ：

rm -rf /etc/containerd/config.toml

systemctl restart containerd

// If execution fails , Reset required

kubeadm reset

// Configure client kubectl

mkdir -p $HOME/.kube

sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

sudo chown $(id -u):$(id -g) $HOME/.kube/config

5.2 initialization worker

//worker Nodes are added to the cluster ---> stay worker node node1 On the implementation

// Can pass kubeadm token list see token

// You can also obtain commands at one time through the following commands

kubeadm token create --print-join-command

5.3 add to pod The Internet calico --> master Node execution

wget https://docs.projectcalico.org/v3.20/manifests/calico.yaml --no-check-certificate

kubectl apply -f calico.yaml

// After application here node State by NotReady --> Ready

5.4 Control surface view cluster

kubectl get nodes

kubectl get pod -A

5.5 The workload worker node node1 View cluster

// stay master01 Transfer documents to node01, among 172.168.186.156 by master01 node

scp -r /etc/kubernetes/admin.conf [email protected]:/etc/kubernetes/

// stay node1 On the implementation

echo "export KUBECONFIG=/etc/kubernetes/admin.conf" >> ~/.bash_profile

source ~/.bash_profile