Kubernetes Cluster Setup: Environment Preparation and Master Node Initialization

Prepare the three servers (master, node‑1, node‑2) by assigning hostnames and IPs, disabling firewalld ( systemctl stop firewalld && systemctl disable firewalld ), turning off SELinux ( sed -i 's/enforcing/disabled/' /etc/selinux/config && setenforce 0 ), and disabling swap ( swapoff -a and editing /etc/fstab ).

Update /etc/hosts with the hostname‑IP mappings and configure bridge traffic to be processed by iptables by creating /etc/sysctl.d/k8s.conf containing: net.bridge.bridge-nf-call-ip6tables = 1 net.bridge.bridge-nf-call-iptables = 1 Then apply with sysctl --system .

Set up Alibaba Cloud YUM mirrors for Docker and Kubernetes: download the Docker repo file to /etc/yum.repos.d/docker-ce.repo and create /etc/yum.repos.d/kubernetes.repo with the appropriate baseurl and GPG keys.

Install Docker ( yum -y install docker-ce-18.06.1.ce-3.el7 ) and enable it ( systemctl enable docker && systemctl start docker ). Install Kubernetes binaries ( yum install -y kubelet-1.19.3 kubeadm-1.19.3 kubectl-1.19.3 ipvsadm ) and configure /etc/sysconfig/kubelet to ignore swap ( KUBELET_EXTRA_ARGS="--fail-swap-on=false" ).

Apply additional kernel parameters ( net.ipv4.ip_forward = 1 ) by adding them to /etc/sysctl.d/k8s.conf and reloading. Enable and start kubelet ( systemctl enable kubelet && systemctl start kubelet ).

Initialize the control plane on the master node with a kubeadm init command, for example: kubeadm init \ --apiserver-advertise-address=192.168.210.85 \ --image-repository=registry.aliyuncs.com/google_containers \ --kubernetes-version=v1.19.3 \ --service-cidr=10.1.0.0/16 \ --pod-network-cidr=10.2.0.0/16 \ --ignore-preflight-errors=NumCPU,Swap This creates certificates, etcd, kube‑apiserver, controller‑manager, scheduler, and writes configuration files.

The guide explains each kubeadm phase (init, preflight, kubelet‑start, certificates, kubeconfig, control‑plane, etcd, wait‑control‑plane, apiclient, uploadconfig, kubelet, patchnode, mark‑control‑plane, bootstrap‑token, add‑ons) and what resources they generate.

After a successful init, copy the admin kubeconfig to a regular user’s $HOME/.kube/config ( mkdir -p $HOME/.kube && cp -i /etc/kubernetes/admin.conf $HOME/.kube/config && chown $(id -u):$(id -g) $HOME/.kube/config ), then deploy a pod network (e.g., Calico) with kubectl apply -f [podnetwork].yaml . Finally, join the worker nodes using the token command shown by kubeadm, such as: kubeadm join 192.168.210.85:6443 --token --discovery-token-ca-cert-hash sha256: .

The article also includes several screenshots illustrating the steps and notes that users should check logs for any errors during the process.