Teach you step by step how to install a K8S cluster with a single master

CKA learning path

---

In the cloud native era, let’s learn together

Article directory

• System environment
• Server planning
• 1. Pre-configuration
• 2. Install Docker and kubeadm/kubelet
• 3. Deploy K8sMaster node
• 4. Join K8sNode node
• 5. Deploy container network (CNI)
• 6. Deploy Dashboard

System environment:

CentOS	7.9
Docker	20-ce
Kubernetes	1.23

Server planning:

192.168.1.186	k8s-master
192.168.1.187	k8s-node1
192.168.1.188	k8s-node2

1. Pre-configuration

1. Execute yum update -y to update the server system to the latest

yum update -y

After successful execution, the expected results are as follows:

 2. Close the system protection wall, Selinux, and Swap partitions (all nodes)

#关闭系统防火墙：
systemctl stop firewalld
systemctl disable firewalld
#关闭Selinux
sed -i 's/enforcing/disabled/' /etc/selinux/config
#关闭Swap分区
sed -ri 's/.*swap.*/#&/' /etc/fstab

The expected results are as follows (the screenshot is only the Master node):

 Add hosts on the Master node

cat >> /etc/hosts << EOF
192.168.1.186 k8s-master
192.168.1.187 k8s-node1
192.168.1.188 k8s-node2
EOF

The expected result is as shown in the figure:

 Pass bridged IPV4 traffic to iptables

cat > /etc/sysctl.d/k8s.conf << EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF

Make the configuration take effect:

sysctl --system

 Expected results after configuration:

2. Install Docker, kubeadm, and kubelet [all nodes]

1. Install Docker

#下载docker的阿里云安装源
wget https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo -O /etc/yum.repos.d/docker-ce.repo
#安装docker
yum -y install docker-ce
#设置为开机启动，并启动docker
systemctl enable docker && systemctl start docker

The expected results are as follows:

 Configure Docker accelerator

#配置docker加速
cat > /etc/docker/daemon.json << EOF
{
  "registry-mirrors": ["https://b9pmyelo.mirror.aliyuncs.com"],
  "exec-opts": ["native.cgroupdriver=systemd"]
}
EOF
#重启docker
systemctl restart docker
#查看docker信息

docker info

The expected results are as follows:

 3. Configure Alibaba Cloud yum software source and install Kubeadm, kubelet, and kubectl

cat > /etc/yum.repos.d/kubernetes.repo << EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=0
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

 4. Install the specified version of K8s, specify version 1.23.0 to install

yum install -y kubelet-1.23.0 kubeadm-1.23.0 kubectl-1.23.0

Installation process:

 Expected results after installation is complete:

 Add startup

systemctl enable kubelet

 

3. Deploy K8SMaster node

Execute on the planned master node

kubeadm init \
  --apiserver-advertise-address=192.168.1.186 \
  --image-repository registry.aliyuncs.com/google_containers \
  --kubernetes-version v1.23.0 \
  --service-cidr=10.96.0.0/12 \
  --pod-network-cidr=10.244.0.0/16 \
  --ignore-preflight-errors=all

Installation process:

---

The expected results after installation are as follows:

 Note: There will be prompts during this step of installation.

[kubelet-check] The HTTP call equal to 'curl -sSL http://localhost:10248/healthz' failed with error: Get "http://localhost:10248/healthz": dial tcp [::1]:10248: connect: connection refused.

    Unfortunately, an error has occurred:
        timed out waiting for the condition

    This error is likely caused by:
        - The kubelet is not running
        - The kubelet is unhealthy due to a misconfiguration of the node in some way (required cgroups disabled)

    If you are on a systemd-powered system, you can try to troubleshoot the error with the following commands:
        - 'systemctl status kubelet'
        - 'journalctl -xeu kubelet'

    Additionally, a control plane component may have crashed or exited when started by the container runtime.
    To troubleshoot, list all containers using your preferred container runtimes CLI.

    Here is one example how you may list all Kubernetes containers running in docker:
        - 'docker ps -a | grep kube | grep -v pause'
        Once you have found the failing container, you can inspect its logs with:
        - 'docker logs CONTAINERID'

error execution phase wait-control-plane: couldn't initialize a Kubernetes cluster
To see the stack trace of this error execute with --v=5 or higher

This is caused by not directly closing the swap partition temporarily. Just execute swapoff -a to restart kubelet.

#临时关闭swap分区
swapoff -a

 4. Join K8Snode node

Use kubeadm to join the node node to the cluster (all child nodes must be executed):

kubeadm join 192.168.1.186:6443 --token rd8ymi.oeztlaletd9agkp0 \
	--discovery-token-ca-cert-hash sha256:b0c8918fc95faab2cc71d49a8f7ebde6902c4e5c60034f91252db6454f01c24c

 Add kubectl environment variables and authentication, and view the added node nodes

mkdir -p $HOME/.kube
cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config
#查看node节点
kubectl get nodes

 The node is currently being prepared and is not running.

The prepared expected results are as shown in the figure:

 Note: The default token is valid for 24 hours. If it exceeds 24 hours, it needs to be regenerated:

kubeadm token create --print-join-command

5. Join the Calico (CNI) container network

calico is a pure three-layer data network center solution, which is a mainstream network solution for k8s.

downloadyaml

wget https://docs.projectcalico.org/manifests/calico.yaml

After downloading, deploy calico

kubectl apply -f calico.yaml

View deployment information of calico

 kubectl get pods -n kube-system

The expected result is as shown in the figure:

 6. Deploy Dashboard

Dashboard is a UI officially provided by k8s, which can manage the resources of the K8S cluster.

vi recommended.yaml
...
kind: Service
apiVersion: v1
metadata:
  labels:
    k8s-app: kubernetes-dashboard
  name: kubernetes-dashboard
  namespace: kubernetes-dashboard
spec:
  ports:
    - port: 443
      targetPort: 8443
      nodePort: 30001
  selector:
    k8s-app: kubernetes-dashboard
  type: NodePort
...

kubectl apply -f recommended.yaml
kubectl get pods -n kubernetes-dashboard

Add to:

nodePort: 30001

type: NodePort

# 创建用户
kubectl create serviceaccount dashboard-admin -n kube-system
# 用户授权
kubectl create clusterrolebinding dashboard-admin --clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin
# 获取用户Token
kubectl describe secrets -n kube-system $(kubectl -n kube-system get secret | awk '/dashboard-admin/{print $1}')

The accessed address is: http://masterip:30001

Use the obtained Token to log in when logging in.

The login expectation is as follows:

Welcome everyone to follow my official account, learn knowledge about operation and maintenance, security, and development together, work hard together, and make progress together.