云服务器 客户端

镜像下载、域名解析、时间同步请点击

阿里巴巴开源镜像站-OPSX镜像站-阿里云开发者社区

1. 服务器要求：

建议最小硬件配置：2核CPU、2G内存、20G硬盘服务器最好可以访问外网，会有从网上拉取镜像需求，如果服务器不能上网，需要提前下载对应镜像并导入节点

1.1 软件环境：

1.2 服务器规划：

1.3 架构图：

2. 操作系统初始化配置

关闭防火墙systemctlstopfirewalldsystemctldisablefirewalld关闭selinuxsed-i/^SELINUX/s/enforcing/disabled//etc/selinux/config永久setenforce0临时关闭swapswapoff-a临时sed-ris/.*swap.*/&//etc/fstab永久根据规划设置主机名hostnamectlset-hostname在master添加hostscat>>/etc/hosts<<EOF192.168.137.81k8s-master1192.168.137.82k8s-node1192.168.137.83k8s-node2EOF将桥接的IPv4流量传递到iptables的链cat>/etc/sysctl.d/k8s.conf<<EOFnet.bridge.bridge-nf-call-ip6tables=1net.bridge.bridge-nf-call-iptables=1EOFsysctl--system生效时间同步yuminstallntpdate-yntpdatetime.windows.com

3. 安装Docker/kubeadm/kubelet [所有节点]

这里使用Docker作为容器引擎，也可以换成别的，例如containerd

3.1 安装Docker

wgethttps://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo -O /etc/yum.repos.d/docker-ce.repoyum-y install docker-cesystemctlenable docker && systemctl start docker

3.2 配置镜像下载加速器

sudomkdir-p /etc/docker sudo tee /etc/docker/daemon.json <<-EOF{"registry-mirrors": ["https://v27k018o.mirror.aliyuncs.com"] } EOF sudo systemctl daemon-reload sudo systemctl restart docker

3.3 添加阿里云YUM软件源

cat> /etc/yum.repos.d/kubernetes.repo << EOF[kubernetes]name=Kubernetesbaseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64enabled=1gpgcheck=0repo_gpgcheck=0gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpgEOF

3.4 安装kubeadm，kubelet，kubectl

指定需安装版本号进行部署

yuminstall-ykubelet-1.20.0kubeadm-1.20.0kubectl-1.20.0systemctlenablekubelet

4. 部署Kubernetes Master

在192.168.137.81（Master）执行

kubeadminit \ --apiserver-advertise-address=192.168.137.81 \ --image-repository registry.aliyuncs.com/google_containers \ --kubernetes-version v1.20.0 \ --service-cidr=10.96.0.0/12 \ --pod-network-cidr=10.244.0.0/16 \ --ignore-preflight-errors=all–apiserver-advertise-address 集群通告地址–image-repository 由于默认拉取镜像地址k8s.gcr.io国内无法访问，这里指定阿里云镜像仓库地址–kubernetes-version K8s版本，与上面安装的一致–service-cidr 集群内部虚拟网络，Pod统一访问入口–pod-network-cidr Pod网络，与下面部署的CNI网络组件yaml中保持一致

或者使用配置文件引导：

cat kubeadm.confapiVersion:kubeadm.k8s.io/v1beta2kind:ClusterConfigurationkubernetesVersion:v1.20.0imageRepository:registry.aliyuncs.com/google_containersnetworking:podSubnet:10.244.0.0/16serviceSubnet:10.96.0.0/12kubeadm init --config kubeadm.conf --ignore-preflight-errors=all

初始化完成后，最后会输出一个join命令，先记住，下面用。

Your Kubernetes control-plane has initialized successfully! Tostartusingyour cluster, you needtorun thefollowingasa regularuser: mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id-u):$(id-g) $HOME/.kube/config Alternatively,ifyouarethe rootuser, you can run:exportKUBECONFIG=/etc/kubernetes/admin.conf You shouldnowdeploy a pod networktothe cluster. Run"kubectl apply -f [podnetwork].yaml"withoneofthe options listedat: https://kubernetes.io/docs/concepts/cluster-administration/addons/Thenyou canjoinanynumberofworker nodesbyrunning thefollowingoneachasroot: kubeadmjoin192.168.137.81:6443--token tix7ff.pnjcwvl6awyaeh8i \--discovery-token-ca-cert-hash sha256:617842ea5040ce5e7f971d387b58693cbfa79261763b68c589fe8d124f1a5154

拷贝kubectl使用的连接k8s认证文件到默认路径：

mkdir-p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudochown$(id -u):$(id -g) $HOME/.kube/config

查看工作节点：

阿里云服务器 vps

[root@k8s-master ~]kubectlgetnodesNAMESTATUSROLESAGEVERSIONk8s-master1NotReadycontrol-plane,master95sv1.20.0

注：由于网络插件还没有部署，还没有准备就绪 NotReady

5. 加入Kubernetes Node节点

在192.168.137.82/83 (Node) 执行 向集群添加新节点，执行在kubeadm init输出的kubeadm join命令：

kubeadmjoin192.168.137.81:6443--tokentix7ff.pnjcwvl6awyaeh8i\--discovery-token-ca-cert-hashsha256:617842ea5040ce5e7f971d387b58693cbfa79261763b68c589fe8d124f1a5154

默认token有效期为24小时，当过期之后，该token就不可用了。这时就需要重新创建token，可以直接使用命令快捷生成：

kubeadm tokencreate--print-join-command

参考资料：https://kubernetes.io/docs/reference/setup-tools/kubeadm/kubeadm-join/

6. 部署容器网络（CNI）

Calico是一个纯三层的数据中心网络方案，是目前Kubernetes主流的网络方案。 下载YAML：

wgethttps://docs.projectcalico.org/manifests/calico.yaml

下载完后还需要修改里面定义Pod网络（CALICO_IPV4POOL_CIDR），与前面kubeadm init的 --pod-network-cidr指定的一样。 修改完后文件后，部署：

kubectlapply -f calico.yamlkubectlget pods -n kube-system

等Calico Pod都Running，节点也会准备就绪：

参考资料：https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/create-cluster-kubeadm/pod-network

[root@k8s-master]kubectlgetpods-nkube-systemNAMEREADYSTATUSRESTARTSAGEcalico-kube-controllers-5f6cfd688c-4hfmv1/1Running02m37scalico-node-8swrp1/1Running02m37scalico-node-nv96p1/1Running02m37scalico-node-s2vl61/1Running02m37scoredns-7f89b7bc75-7nzm80/1Running022mcoredns-7f89b7bc75-9c46c0/1Running022metcd-k8s-master11/1Running022mkube-apiserver-k8s-master11/1Running022mkube-controller-manager-k8s-master11/1Running022mkube-proxy-clc6k1/1Running018mkube-proxy-dvlvr1/1Running022mkube-proxy-jr6hm1/1Running018mkube-scheduler-k8s-master11/1Running022m[root@k8s-master]kubectlgetnodesNAMESTATUSROLESAGEVERSIONk8s-master1Readycontrol-plane,master23mv1.20.0k8s-node1Ready18mv1.20.0k8s-node2Ready18mv1.20.0

7. 测试kubernetes集群

在Kubernetes集群中创建一个pod，验证是否正常运行：

kubectlcreatedeployment nginx--image=nginxkubectl expose deployment nginx--port=80 --type=NodePortkubectlgetpod,svc[root@k8s-master~]kubectlgetpod,svcNAMEREADYSTATUSRESTARTSAGEpod/nginx-6799fc88d8-nkzk61/1Running06m51sNAMETYPECLUSTER-IPEXTERNAL-IPPORT(S)AGEservice/kubernetesClusterIP10.96.0.1443/TCP32mservice/nginxNodePort10.108.121.25280:30167/TCP6m2s

访问地址：http://NodeIP:Port

阿里云esc服务器

8. 部署Dashboard

Dashboard是官方提供的一个UI，可用于基本管理K8s资源。

wgethttps://raw.githubusercontent.com/kubernetes/dashboard/v2.0.3/aio/deploy/recommended.yaml

默认Dashboard只能集群内部访问，修改Service为NodePort类型，暴露到外部：

vi recommended.yaml...kind:ServiceapiVersion:v1metadata:labels:k8s-app:kubernetes-dashboardname:kubernetes-dashboardnamespace:kubernetes-dashboardspec:ports:-port:443targetPort:8443nodePort:30001selector:k8s-app:kubernetes-dashboardtype:NodePort...kubectl apply -f recommended.yamlkubectl get pods -n kubernetes-dashboard

访问地址：https://NodeIP:30001 创建service account并绑定默认cluster-admin管理员集群角色：

创建用户$ kubectlcreateserviceaccount dashboard-admin-n kube-system用户授权$ kubectlcreateclusterrolebinding dashboard-admin--clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin获取用户Token$ kubectldescribesecrets -n kube-system$(kubectl -n kube-systemgetsecret | awk/dashboard-admin/{print $1})

使用输出的token登录Dashboard。

[root@k8s-master~]kubectldescribesecrets-nkube-system$(kubectl-nkube-systemgetsecret|awk/dashboard-admin/{print $1})Name:dashboard-admin-token-cgmldNamespace:kube-systemLabels:Annotations: kubernetes.io/service-account.name:dashboard-adminkubernetes.io/service-account.uid:3208e176-5115-4425-bb67-d45863bc05f7Type:kubernetes.io/service-account-tokenData====ca.crt:1066bytesnamespace:11bytestoken:eyJhbGciOiJSUzI1NiIsImtpZCI6IlBnN2lzMk8zYndPX3ZONnc0cnFnRjhsVnczOTVlNGxXSDl4c1Z0OGtmNDAifQ.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.DY9SLeakUzVUqUHLJjLuaBtv0EOj6l-zsCfmzKTtsiTkaX39bGpLInToKuSpbXHYAmkpCvoZH22ghmOds3LFgvQBCIt6M83rrL83aPzhDjAKtPtPkz9vJGR7K5LnfrB9AX5dVhiU_AkaVIBIFcqTxVlFpl6W1EzTc0uJDM7K8Gr2XnPvRfUMe8WaEWR7tVxMEEhPhP2waEYmcc5uFz5unI_g6lTMYRJnhZCjfqh7lS9NA_8WgmoQnQjxW4cYAsqrdCzbroTEMCslH_pCj-PZNxf7mKVXZwklYL78t8klU_AytuhdaV88iRR3HEuBMYLbfJjy6RLkyt_ORweaXb8npg

本文转自：https://blog.51cto.com/wemux/5354074

云服务器无法连接