Prepare k8s Cluster

Building a K8s Cluster, you can choose one of the following methods.

Install Kuberctl

Build Cluster

Install By

Prerequisites

  • Hardware Requirements:

    1. At least 2 GB of RAM per machine (minimum 1 GB)
    2. 2 CPUs on the master node
    3. Full network connectivity among all machines (public or private network)

  • Operating System:

    1. Ubuntu 20.04/18.04, CentOS 7/8, or any other supported Linux distribution.

  • Network Requirements:

    1. Unique hostname, MAC address, and product_uuid for each node.
    2. Certain ports need to be open (e.g., 6443, 2379-2380, 10250, 10251, 10252, 10255, etc.)

  • Disable Swap:

    sudo swapoff -a

Steps to Setup Kubernetes Cluster

  1. Prepare Your Servers Update the Package Index and Install Necessary Packages On all your nodes (both master and worker):
sudo apt-get update
sudo apt-get install -y apt-transport-https ca-certificates curl

Add the Kubernetes APT Repository

curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | sudo apt-key add -
cat <<EOF | sudo tee /etc/apt/sources.list.d/kubernetes.list
deb http://apt.kubernetes.io/ kubernetes-xenial main
EOF

Install kubeadm, kubelet, and kubectl

sudo apt-get update
sudo apt-get install -y kubelet kubeadm kubectl
sudo apt-mark hold kubelet kubeadm kubectl
  1. Initialize the Master Node On the master node, initialize the Kubernetes control plane:
sudo kubeadm init --pod-network-cidr=192.168.0.0/16

The –pod-network-cidr flag is used to set the Pod network range. You might need to adjust this based on your network provider

Set up Local kubeconfig

mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
  1. Install a Pod Network Add-on You can install a network add-on like Flannel, Calico, or Weave. For example, to install Calico:
  1. Join Worker Nodes to the Cluster On each worker node, run the kubeadm join command provided at the end of the kubeadm init output on the master node. It will look something like this:
sudo kubeadm join <master-ip>:6443 --token <token> --discovery-token-ca-cert-hash sha256:<hash>

If you lost the join command, you can create a new token on the master node:

sudo kubeadm token create --print-join-command
  1. Verify the Cluster Once all nodes have joined, you can verify the cluster status from the master node:
kubectl get nodes

This command should list all your nodes with the status “Ready”.

Subsections of Prepare k8s Cluster

Kind

Preliminary

  • Kind binary has installed, if not check 🔗link

  • Hardware Requirements:

    1. At least 2 GB of RAM per machine (minimum 1 GB)
    2. 2 CPUs on the master node
    3. Full network connectivity among all machines (public or private network)

  • Operating System:

    1. Ubuntu 22.04/14.04, CentOS 7/8, or any other supported Linux distribution.

  • Network Requirements:

    1. Unique hostname, MAC address, and product_uuid for each node.
    2. Certain ports need to be open (e.g., 6443, 2379-2380, 10250, 10251, 10252, 10255, etc.)

Customize your cluster

Creating a Kubernetes cluster is as simple as kind create cluster

kind create cluster --name test

Reference

and the you can visit https://kind.sigs.k8s.io/docs/user/quick-start/ for mode detail.

K3s

Preliminary

  • Hardware Requirements:

    1. Server need to have at least 2 cores, 2 GB RAM
    2. Agent need 1 core , 512 MB RAM

  • Operating System:

    1. K3s is expected to work on most modern Linux systems.

  • Network Requirements:

    1. The K3s server needs port 6443 to be accessible by all nodes.
    2. If you wish to utilize the metrics server, all nodes must be accessible to each other on port 10250.

Init server

curl -sfL https://rancher-mirror.rancher.cn/k3s/k3s-install.sh | INSTALL_K3S_MIRROR=cn sh -s - server --cluster-init --flannel-backend=vxlan --node-taint "node-role.kubernetes.io/control-plane=true:NoSchedule"

Get token

cat /var/lib/rancher/k3s/server/node-token

Join worker

curl -sfL https://rancher-mirror.rancher.cn/k3s/k3s-install.sh | INSTALL_K3S_MIRROR=cn K3S_URL=https://<master-ip>:6443 K3S_TOKEN=<join-token> sh -

Copy kubeconfig

mkdir -p $HOME/.kube
cp /etc/rancher/k3s/k3s.yaml $HOME/.kube/config

Minikube

Preliminary

  • Minikube binary has installed, if not check 🔗link

  • Hardware Requirements:

    1. At least 2 GB of RAM per machine (minimum 1 GB)
    2. 2 CPUs on the master node
    3. Full network connectivity among all machines (public or private network)

  • Operating System:

    1. Ubuntu 20.04/18.04, CentOS 7/8, or any other supported Linux distribution.

  • Network Requirements:

    1. Unique hostname, MAC address, and product_uuid for each node.
    2. Certain ports need to be open (e.g., 6443, 2379-2380, 10250, 10251, 10252, 10255, etc.)

[Optional] Disable aegis service and reboot system for Aliyun

sudo systemctl disable aegis && sudo reboot

Customize your cluster

minikube start --driver=podman  --image-mirror-country=cn --kubernetes-version=v1.33.1 --image-repository=registry.cn-hangzhou.aliyuncs.com/google_containers --cpus=6 --memory=20g --disk-size=50g --force

Restart minikube

minikube stop && minikube start

Add alias

alias kubectl="minikube kubectl --"

Forward

ssh -i ~/.minikube/machines/minikube/id_rsa docker@$(minikube ip) -L '*:30443:0.0.0.0:30443' -N -f

and then you can visit https://minikube.sigs.k8s.io/docs/start/ for more detail.

FAQ

通常是由于 Metrics Server 未正确安装 或 External Metrics API 缺失 导致的

# 启用 Minikube 的 metrics-server 插件
minikube addons enable metrics-server

# 等待部署完成（约 1-2 分钟）
kubectl wait --for=condition=available deployment/metrics-server -n kube-system --timeout=180s

# 验证 Metrics Server 是否运行
kubectl -n kube-system get pods  | grep metrics-server

the possibilities are endless (almost - including other shortcodes may or may not work)

You can add standard markdown syntax:

  • multiple paragraphs
  • bullet point lists
  • emphasized, bold and even bold emphasized text
  • links
  • etc.
...and even source code

the possibilities are endless (almost - including other shortcodes may or may not work)