跳到主要内容

Sealos 部署 Kubernetes 集群

最后于 更新

本文档用于在虚机环境中使用 Sealos 快速部署 Kubernetes 集群单节点环境。

准备工作​

  • 单台 8C16G 纯净虚机。
  • 建议使用干净的操作系统来创建集群。不要自己装 Docker！
  • 支持大多数 Linux 发行版，例如：Ubuntu、CentOS、Rocky linux。
  • 支持使用 Containerd 作为容器运行时。
  • 一定需要 root 用户部署

联网环境部署 Sealos Kubernetes​

mkdir /opt/sealos
cd /opt/sealos
curl http://storage.kindlingx.com/package/sealos-k8s/sealos-k8s-deploy-v1.1.0-online.sh | bash

离线环境部署 Sealos Kubernetes​

下载离线安装包

mkdir ./sealos
cd ./sealos
curl http://storage.kindlingx.com/package/sealos-k8s/sealos-k8s-package-download-v1.1.0.sh | bash

执行下方命令检查安装包

ls -l # ~/sealos
total 951232
-rw-r--r--. 1 root root 351843328 Aug 16 16:02 calico-v3.24.1.tar
-rw-r--r--. 1 root root  17112576 Aug 16 16:00 helm-v3.14.0.tar
-rw-r--r--. 1 root root 411297280 Aug 16 16:01 kubernetes-v1.25-latest.tar
-rw-r--r--. 1 root root 130833408 Aug 16 16:02 openebs-v3.10.0.tar
-rw-r--r--. 1 root root      1175 Aug 16 16:02 sealos-k8s-deploy-v1.1.0.sh
-rw-r--r--. 1 root root  62963267 Aug 16 16:00 sealos_4.3.7_linux_amd64.tar.gz

将sealos文件夹下所有文件移动到目标部署机器的/opt/sealos文件夹下，并在/opt/sealos文件夹下执行命令：

bash sealos-k8s-deploy-v1.1.0.sh

验证​

脚本执行完成后预期有类似输出

2024-08-16T16:19:57 info succeeded in creating a new cluster, enjoy it!
2024-08-16T16:19:57 info
      ___           ___           ___           ___       ___           ___
     /\  \         /\  \         /\  \         /\__\     /\  \         /\  \
    /::\  \       /::\  \       /::\  \       /:/  /    /::\  \       /::\  \
   /:/\ \  \     /:/\:\  \     /:/\:\  \     /:/  /    /:/\:\  \     /:/\ \  \
  _\:\~\ \  \   /::\~\:\  \   /::\~\:\  \   /:/  /    /:/  \:\  \   _\:\~\ \  \
 /\ \:\ \ \__\ /:/\:\ \:\__\ /:/\:\ \:\__\ /:/__/    /:/__/ \:\__\ /\ \:\ \ \__\
 \:\ \:\ \/__/ \:\~\:\ \/__/ \/__\:\/:/  / \:\  \    \:\  \ /:/  / \:\ \:\ \/__/
  \:\ \:\__\    \:\ \:\__\        \::/  /   \:\  \    \:\  /:/  /   \:\ \:\__\
   \:\/:/  /     \:\ \/__/        /:/  /     \:\  \    \:\/:/  /     \:\/:/  /
    \::/  /       \:\__\         /:/  /       \:\__\    \::/  /       \::/  /
     \/__/         \/__/         \/__/         \/__/     \/__/         \/__/

                  Website: https://www.sealos.io/
                  Address: github.com/labring/sealos
                  Version: 4.3.7-f39b2339

# 删除控制节点污点，报错不影响最终结果
error: taint "node-role.kubernetes.io/master" not found
error: taint "node-role.kubernetes.io/control-plane" not found

执行下方命令等待并确认所有 POD 处于Running状态

[root@demo sealos]# kubectl get po -A
NAMESPACE          NAME                                           READY   STATUS    RESTARTS   AGE
calico-apiserver   calico-apiserver-858b677f96-qlh9v              1/1     Running   0          4m24s
calico-apiserver   calico-apiserver-858b677f96-rm6js              1/1     Running   0          4m24s
calico-system      calico-kube-controllers-85666c5b94-rkp4v       1/1     Running   0          5m22s
calico-system      calico-node-xnzz2                              1/1     Running   0          5m22s
calico-system      calico-typha-67b6658b4f-49rmq                  1/1     Running   0          5m22s
calico-system      csi-node-driver-gbhdz                          2/2     Running   0          4m52s
kube-system        coredns-565d847f94-6ghzd                       1/1     Running   0          5m31s
kube-system        coredns-565d847f94-cqpxm                       1/1     Running   0          5m31s
kube-system        etcd-node-69                                   1/1     Running   2          5m43s
kube-system        kube-apiserver-node-69                         1/1     Running   2          5m43s
kube-system        kube-controller-manager-node-69                1/1     Running   2          5m43s
kube-system        kube-proxy-56wwf                               1/1     Running   0          5m31s
kube-system        kube-scheduler-node-69                         1/1     Running   2          5m46s
openebs            openebs-localpv-provisioner-66454cfbb9-99bhd   1/1     Running   0          5m31s
tigera-operator    tigera-operator-6675dc47f4-6b4fm               1/1     Running   0          5m31s

卸载 Sealos Kubernetes​

sealos reset