第零章（K8s启航）：最新Ubuntu25 安装最新K8S （断电重启、断电重置）超详细步骤，安装不好你来打我~

Ubuntu安装K8S1. 服务器初始化所有节点vim/etc/hosts127.0.0.1 localhost # 127.0.1.1 yww # The following lines are desirable for IPv6 capable hosts ::1 ip6-localhost ip6-loopback fe00::0 ip6-localnet ff00::0 ip6-mcastprefix ff02::1 ip6-allnodes ff02::2 ip6-allrouters 192.168.163.3 k8s-master 192.168.163.4 k8s-node1 192.168.163.5 k8s-node2hostnamectl set-hostname k8s-master hostnamectl set-hostname k8s-node1 hostnamectl set-hostname k8s-node2sudoaptupdatesudotimedatectl set-timezone Asia/Shanghaisudoaptinstallntpsec-ntpdate# 同步一sudontpdate time.windows.com# 同步二sudontpdate pool.ntp.org禁用swap分区//1方式 // 注释掉 /etc/fstab 中的 Swap 条目 // 备份 fstab 文件sudocp/etc/fstab /etc/fstab.bak //编辑 fstab 文件sudovim/etc/fstab //在文件中找到包含 swap 的行在行首添加# 注释掉:示例: /dev/disk/by-uuid/63c40f44-82d7-4ae8-9e1c-e85209d91664 / ext4 defaults01# /swap.img none swap sw 0 0修改linux的内核参数sudomodprobe br_netfiltersudosysctlnet.bridge.bridge-nf-call-iptables1配置ipvs功能//1. 安装必要的软件包,更新系统sudoaptupdate // 安装 ipset 和 ipvsadmsudoaptinstallipset ipvsadm-y// 验证安装 ipset--versionipvsadm--version//2. 加载 IPVS 内核模块 // 加载 IPVS 相关内核模块sudomodprobe ip_vssudomodprobe ip_vs_rrsudomodprobe ip_vs_wrrsudomodprobe ip_vs_shsudomodprobe nf_conntrack // 验证模块是否加载 lsmod|grepip_vs重启reboot2. 安装docker所有节点// 设置 Docker 的apt仓库 //1. 添加 Docker 官方的 GPG 密钥sudoapt-getupdatesudoapt-getinstallca-certificatescurlsudoinstall-m0755-d/etc/apt/keyringssudocurl-fsSLhttps://download.docker.com/linux/ubuntu/gpg-o/etc/apt/keyrings/docker.ascsudochmodar /etc/apt/keyrings/docker.asc //2. 将存储库添加到 Apt 源echo\deb [arch$(dpkg --print-architecture)signed-by/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu \$(./etc/os-releaseecho${UBUNTU_CODENAME:-$VERSION_CODENAME})stable|\sudotee/etc/apt/sources.list.d/docker.list/dev/nullsudoapt-getupdate //3. 安装 Docker 软件包。sudoapt-getinstalldocker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin //4. 配置镜像源sudonano/etc/docker/daemon.json{registry-mirrors:[https://docker.m.daocloud.io,https://dockerproxy.com,https://docker.mirrors.ustc.edu.cn,https://docker.nju.edu.cn,http://hub-mirror.c.163.com],log-driver:json-file,log-opts:{max-size:200m,max-file:5}}5. 重启dockersudosystemctl daemon-reloadsudosystemctl restartdocker6. 检查配置是否生效dockerinfodockerinfo|grep-ECgroup|Loggingdockerinfo|grep-A5Registry Mirrors# Cgroup Driver: systemd# Logging Driver: json-file8. 测试dockersudodockerrun hello-world避免 docker containerd 使用系统盘可忽略# /var/lib/docker 已经是数据盘# /var/lib/containerd --- 数据盘# 停服务systemctl stopdockersystemctl stop containerd# 创建目录mkdir-p/var/lib/docker/containerd# 迁移数据mv/var/lib/containerd /var/lib/docker/containerd/# 做软链接ln-s/var/lib/docker/containerd/containerd /var/lib/containerd# 启动服务systemctl daemon-reexec systemctl restart containerd systemctl restartdocker3. 安装 cri-docker所有节点//1. 安装依赖sudoaptupdatesudoaptinstall-ycurlwget//2. 下载 cri-dockerd 二进制文件 // 获取最新版本号CRI_DOCKERD_VERSION$(curl-shttps://api.github.com/repos/Mirantis/cri-dockerd/releases/latest|greptag_name:|sed-Es/.*([^]).*/\1/)//3. 下载并解压wgethttps://github.com/Mirantis/cri-dockerd/releases/download/${CRI_DOCKERD_VERSION}/cri-dockerd-${CRI_DOCKERD_VERSION#v}.amd64.tgztarxvf cri-dockerd-${CRI_DOCKERD_VERSION#v}.amd64.tgz //4. 安装到系统路径sudomvcri-dockerd/cri-dockerd /usr/local/bin/sudochmodx /usr/local/bin/cri-dockerd //5. 创建新的服务文件sudotee/etc/systemd/system/cri-docker.service/dev/nullEOF [Unit] DescriptionCRI Interface for Docker Application Container Engine Documentationhttps://docs.mirantis.com Afternetwork-online.target docker.service Wantsnetwork-online.target [Service] Typenotify ExecStart/usr/local/bin/cri-dockerd\\--network-plugincni\\--pod-infra-container-imageregistry.k8s.io/pause:3.9\\--container-runtime-endpointunix:///var/run/cri-dockerd.sock ExecReload/bin/kill -s HUP \$MAINPIDTimeoutSec0 RestartSec2 Restartalways StartLimitBurst3 StartLimitInterval60s LimitNOFILEinfinity LimitNPROCinfinity LimitCOREinfinity TasksMaxinfinity Delegateyes KillModeprocess [Install] WantedBymulti-user.target EOF//6. 重新加载 systemdsudosystemctl daemon-reload //7. 启用并启动服务sudosystemctlenablecri-docker.servicesudosystemctl start cri-docker.service //8. 检查状态sudosystemctl status cri-docker.service //9. 检查 socket 文件是否创建ls-la/var/run/cri-dockerd.sock4. 安装kubernetes组件所有节点//1. 更新镜像源sudoaptupdatesudoaptupgrade-ysudoaptinstall-ycurlgnupg apt-transport-https ca-certificates //2. 添加 Kubernetes 国内镜像源阿里云curl-fsSLhttps://mirrors.aliyun.com/kubernetes/apt/doc/apt-key.gpg|sudogpg--dearmor-o/etc/apt/keyrings/kubernetes-aliyun.gpg //3. 添加 Kubernetes 源echodeb [signed-by/etc/apt/keyrings/kubernetes-aliyun.gpg] https://mirrors.aliyun.com/kubernetes/apt/ kubernetes-xenial main|sudotee/etc/apt/sources.list.d/kubernetes.list //4. 更新包列表并安装sudoaptupdate //5. 安装指定版本的 kubeadm、kubelet、kubectl //6. 查看可用版本apt-cachemadison kubeadm //7. 安装最新稳定版本sudoaptinstall-ykubelet kubeadm kubectl //8. 或者安装特定版本推荐# sudo apt install -y kubelet1.28.0-00 kubeadm1.28.0-00 kubectl1.28.0-00//9. 锁定版本防止自动更新sudoapt-mark hold kubelet kubeadm kubectl //10. 配置K8S使用clidockersudomkdir-p/etc/systemd/system/kubelet.service.d //11. 创建 kubelet 服务配置sudotee/etc/systemd/system/kubelet.service.d/0-cri-docker.conf/dev/nullEOF [Service] EnvironmentKUBELET_EXTRA_ARGS--container-runtime-endpointunix:///var/run/cri-dockerd.sock EOF# 拉 pause 镜像sudodockerpull registry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.9# 给 kubelet 用的镜像打 tagsudodockertag registry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.9 k8s.gcr.io/pause:3.9sudodockertag k8s.gcr.io/pause:3.9 registry.k8s.io/pause:3.9vim/etc/systemd/system/kubelet.service.d/10-kubeadm.conf# Note: This dropin only works with kubeadm and kubelet v1.11 [Service] # kubeconfig 文件路径 EnvironmentKUBELET_KUBECONFIG_ARGS--bootstrap-kubeconfig/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig/etc/kubernetes/kubelet.conf EnvironmentKUBELET_CONFIG_ARGS--config/var/lib/kubelet/config.yaml # kubeadm 动态生成的参数 EnvironmentFile-/var/lib/kubelet/kubeadm-flags.env # 用户自定义参数文件 EnvironmentFile-/etc/default/kubelet # 指定 pause 镜像使用本地或国内源 # 如果你的本地已经有 registry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.9 # 或者你已经打了 tag 对应 registry.k8s.io/pause:3.9可以改为你本地镜像 EnvironmentKUBELET_EXTRA_ARGS--pod-infra-container-imageregistry.cn-hangzhou.aliyuncs.com/google_containers/pause:3.9 # 清空 ExecStart使用自定义参数 ExecStart ExecStart/usr/bin/kubelet $KUBELET_KUBECONFIG_ARGS $KUBELET_CONFIG_ARGS $KUBELET_KUBEADM_ARGS $KUBELET_EXTRA_ARGS # 重启策略 Restartalways StartLimitInterval0 RestartSec10sudosystemctl daemon-reload5. 安装kubernetes组件(仅master节点执行)//1. 正确的镜像拉取脚本,修改为你下载的版本images(kube-apiserver:v1.28.2 kube-controller-manager:v1.28.2 kube-scheduler:v1.28.2 kube-proxy:v1.28.2 pause:3.9 etcd:3.5.9-0 coredns:v1.10.1)//2. 进行下载forimageNamein${images[]};doechoPulling$imageNamesudodockerpull registry.aliyuncs.com/google_containers/$imageNamesudodockertag registry.aliyuncs.com/google_containers/$imageNameregistry.k8s.io/$imageNamesudodockerrmi registry.aliyuncs.com/google_containers/$imageNamedone//3. 验证镜像sudodockerimages|grepk8s.io6. 集群初始化(仅master节点执行)sudomodprobe br_netfilterechobr_netfilter|sudotee/etc/modules-load.d/br_netfilter.confsudosysctlnet.bridge.bridge-nf-call-iptables1sudosysctl-pcat/proc/sys/net/bridge/bridge-nf-call-iptables# 如果返回 1 就正常了。sudokubeadm init\--cri-socket unix:///var/run/cri-dockerd.sock\--image-repository registry.aliyuncs.com/google_containers\--kubernetes-version v1.28.2\--service-cidr10.96.0.0/12\--pod-network-cidr10.244.0.0/16# 创建必要文件mkdir-p$HOME/.kubesudocp-i/etc/kubernetes/admin.conf$HOME/.kube/configsudochown$(id-u):$(id-g)$HOME/.kube/config# 日志查看sudojournalctl-ukubelet-f7. 集群初始化node节点sudomodprobe br_netfilterechobr_netfilter|sudotee/etc/modules-load.d/br_netfilter.confsudosysctlnet.bridge.bridge-nf-call-iptables1sudosysctl-pcat/proc/sys/net/bridge/bridge-nf-call-iptablessudokubeadmjoin192.168.163.3:6443\--tokenod0g5i.r538ydrdho8nbybs\--discovery-token-ca-cert-hash sha256:324358e41cd704c75a70ae1111e96f245013b8e244faa622096a96585aee306b\--cri-socket unix:///var/run/cri-dockerd.sock8. 安装网络插件master节点kubectl apply-fhttps://raw.githubusercontent.com/flannel-io/flannel/master/Documentation/kube-flannel.yml至此所有K8S安装步骤完成可以使用kubectl get nodes等等所有节点 Ready 即可。等待所有节点 Ready 需要一点时间耐心等待安装失败或者节点始终 NotReady一定是步骤漏输或输错实在不行使用 断电重置方法 重置之后在初始化 k8s 集群。9. 断电重置master# 停止 kubelet 和 docker/crisudosystemctl stop kubeletsudosystemctl stopdocker# 或者 containerd/cri-o# 重置 kubeadm 配置sudokubeadm reset-f--cri-socket unix:///var/run/cri-dockerd.sock# 删除 CNI 网络插件配置sudorm-rf/etc/cni/net.dsudorm-rf/etc/kubernetes/manifestssudorm-rf/var/lib/etcdsudorm-rf/var/lib/kubelet/*sudorm-rf/etc/kubernetes/pki# 清理 kubelet、kube-proxy 数据sudorm-rf/var/lib/kube-proxysudorm-rf/var/lib/calico /var/lib/flannel# 如果安装了# 重启 docker 和 kubeletsudosystemctl restartdockersudosystemctl restart kubeletsudokubeadm init\--cri-socket unix:///var/run/cri-dockerd.sock\--image-repository registry.aliyuncs.com/google_containers\--kubernetes-version v1.28.2\--service-cidr10.96.0.0/12\--pod-network-cidr10.244.0.0/16mkdir-p$HOME/.kubesudocp-i/etc/kubernetes/admin.conf$HOME/.kube/configsudochown$(id-u):$(id-g)$HOME/.kube/config kubectl apply-fhttps://raw.githubusercontent.com/flannel-io/flannel/master/Documentation/kube-flannel.ymlnode节点sudokubeadm reset-f--cri-socket unix:///var/run/cri-dockerd.socksudorm-rf/etc/kubernetes/*sudorm-rf/var/lib/kubelet/*sudorm-rf/etc/cni/net.d/*sudorm-rf/var/lib/cni/*sudorm-rf/run/flannel/*sudosystemctl stop kubeletsudofuser-k10250/tcp||truesudosystemctl daemon-reloadsudosystemctl restart kubeletsudokubeadmjoin192.168.163.3:6443\--tokeng1ky1i.amhisjcq62fivryb\--discovery-token-ca-cert-hash sha256:975d7ed132348ec49093391eac632cd546996df382e86694882b6f229a33cb28\--cri-socket unix:///var/run/cri-dockerd.sock10. 断电重启所有节点# dockersudosystemctl startdockersudo# 检查 kubeletsudosystemctl status kubeletsudosystemctl start kubelet11 部署nginx进行测试kubectl create deployment nginx--imagenginx:1.14-alpine kubectl expose deployment nginx--port80--typeNodePort kubectl get pods,service ---------------------------------------------------------------------------------- NAME READY STATUS RESTARTS AGE pod/nginx-596ff96db7-8cms51/1 Running030s NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S)AGE service/kubernetes ClusterIP10.96.0.1none443/TCP 16m service/nginx NodePort10.105.192.235none80:30613/TCP 5s ---------------------------------------------------------------------------------- kubectl delete deployment nginx kubectl deleteservicenginx