Ubuntu PC 1台で kubeadm を使って Kubernetes クラスターを構築
Ubuntu PC 1台の上で Multipass を使い、仮想マシン（VM）上に Kubernetes クラスターを構築する。 構成は Master 1台 + Worker 3台（計4台）、kubeadm を使用し、ネットワークプラグインは Calico を採用。 コンテナランタイムとして Docker を利用する。
ref. Calico - IBM Documentation
Ubuntu PC
# CPU
haruki@node01:~$ nproc
4
# メモリ
haruki@node01:~$ free -g
total used free shared buff/cache available
Mem: 11 0 10 0 0 10
Swap: 1 0 1
# ストレージ
haruki@node01:~$ df -Th /
Filesystem Type Size Used Avail Use% Mounted on
/dev/sda2 ext4 468G 20G 425G 5% /
Multipass のインストール
haruki@node01:~$ sudo snap install multipass multipass 1.15.1 from Canonical✓ installed haruki@node01:~$ multipass version multipass 1.15.1 multipassd 1.15.1 haruki@node01:~$ multipass list No instances found.
仮想マシン（VM）の作成
| ノード名 | CPU | メモリ | ストレージ |
|---|---|---|---|
| k8s-master | 2コア | 4GB | 20GB |
| k8s-worker-01 | 1コア | 2GB | 10GB |
| k8s-worker-02 | 1コア | 2GB | 10GB |
| k8s-worker-03 | 1コア | 2GB | 10GB |
haruki@node01:~$ multipass launch --name k8s-master --cpus 2 --mem 4G --disk 20G Warning: the "--mem" long option is deprecated in favour of "--memory". Please update any scripts, etc. Launched: k8s-master haruki@node01:~$ multipass launch --name k8s-worker-01 --cpus 1 --mem 2G --disk 10G Warning: the "--mem" long option is deprecated in favour of "--memory". Please update any scripts, etc. Launched: k8s-worker-01 haruki@node01:~$ multipass launch --name k8s-worker-02 --cpus 1 --mem 2G --disk 10G Warning: the "--mem" long option is deprecated in favour of "--memory". Please update any scripts, etc. Launched: k8s-worker-02 haruki@node01:~$ multipass launch --name k8s-worker-03 --cpus 1 --mem 2G --disk 10G Warning: the "--mem" long option is deprecated in favour of "--memory". Please update any scripts, etc. Launched: k8s-worker-03 haruki@node01:~$ multipass list Name State IPv4 Image k8s-master Running 10.105.167.32 Ubuntu 24.04 LTS k8s-worker-01 Running 10.105.167.224 Ubuntu 24.04 LTS k8s-worker-02 Running 10.105.167.63 Ubuntu 24.04 LTS k8s-worker-03 Running 10.105.167.9 Ubuntu 24.04 LTS
各 Node に Docker & Kubernetes をインストール
# kubelet がメモリ管理をする際、スワップが有効だと挙動が不安定になることがある ubuntu@k8s-master:~$ sudo swapoff -a ubuntu@k8s-master:~$ sudo sed -i '/ swap / s/^/#/' /etc/fstab
Docker インストール
ubuntu@k8s-master:~$ sudo apt update && sudo apt install -y docker.io ubuntu@k8s-master:~$ sudo systemctl enable docker ubuntu@k8s-master:~$ sudo systemctl start docker ubuntu@k8s-master:~$ sudo systemctl status docker
cgroupdriver=systemd にすることで、kubelet が Docker のリソース管理と統一した方法で動作
ubuntu@k8s-master:~$ cat <<EOF | sudo tee /etc/docker/daemon.json
{
"exec-opts": ["native.cgroupdriver=systemd"]
}
EOF
sudo systemctl restart docker
Ubuntu 24.04（Noble）は、Ubuntu 22.04（Jammy）と互換性
ubuntu@k8s-master:~$ sudo apt update ubuntu@k8s-master:~$ sudo apt install -y apt-transport-https ca-certificates curl ubuntu@k8s-master:~$ curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.30/deb/Release.key | sudo tee /etc/apt/trusted.gpg.d/kubernetes.asc ubuntu@k8s-master:~$ echo "deb https://pkgs.k8s.io/core:/stable:/v1.30/deb/ /" | sudo tee /etc/apt/sources.list.d/kubernetes.list ubuntu@k8s-master:~$ sudo apt update
Kubernetes のインストール
ubuntu@k8s-master:~$ sudo apt install -y kubelet kubeadm kubectl ubuntu@k8s-master:~$ sudo systemctl enable kubelet
Master Node で Kubernetes クラスターを作成
# Worker Node を追加するための kubeadm join コマンドが表示される
ubuntu@k8s-master:~$ sudo kubeadm init --pod-network-cidr=192.168.0.0/16
I0310 18:35:35.035243 4576 version.go:256] remote version is much newer: v1.32.2; falling back to: stable-1.30
--- snip ---
Your Kubernetes control-plane has initialized successfully!
To start using your cluster, you need to run the following as a regular user:
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Alternatively, if you are the root user, you can run:
export KUBECONFIG=/etc/kubernetes/admin.conf
You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
https://kubernetes.io/docs/concepts/cluster-administration/addons/
Then you can join any number of worker nodes by running the following on each as root:
sudo kubeadm join <master-ip>:6443 --token <token> \
--discovery-token-ca-cert-hash sha256:<hash>
ubuntu@k8s-master:~$ mkdir -p $HOME/.kube
ubuntu@k8s-master:~$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
ubuntu@k8s-master:~$ sudo chown $(id -u):$(id -g) $HOME/.kube/config
# 2025/03/10 時点の latest
ubuntu@k8s-master:~$ kubectl apply -f https://raw.githubusercontent.com/projectcalico/calico/v3.29.2/manifests/calico.yaml
Worker Node を Kubernetes クラスターに追加
sudo kubeadm join <master-ip>:6443 --token <token> --discovery-token-ca-cert-hash sha256:<hash>
ubuntu@k8s-master:~$ kubectl get nodes NAME STATUS ROLES AGE VERSION k8s-master Ready control-plane 84m v1.30.10 k8s-worker-01 Ready <none> 69m v1.30.10 k8s-worker-02 Ready <none> 35m v1.30.10 k8s-worker-03 Ready <none> 34m v1.30.10
お掃除
いろいろ遊び終わったら削除
haruki@node01:~$ multipass stop k8s-master k8s-worker-01 k8s-worker-02 k8s-worker-03 haruki@node01:~$ multipass delete k8s-master k8s-worker-01 k8s-worker-02 k8s-worker-03 haruki@node01:~$ multipass list Name State IPv4 Image k8s-master Deleted -- Ubuntu 24.04 LTS k8s-worker-01 Deleted -- Ubuntu 24.04 LTS k8s-worker-02 Deleted -- Ubuntu 24.04 LTS k8s-worker-03 Deleted -- Ubuntu 24.04 LTS haruki@node01:~$ multipass purge haruki@node01:~$ multipass list No instances found. haruki@node01:~$
