在生产环境中，不推荐使用，因为没有冗余
前提条件：已经安装好了kuber，请安装前篇k8s安装操作
架构：
在第一台的基础上复制5台机器，配置好环境
1、在master节点:10.21.213.221配置k8s配置文件
cat /etc/kubernetes/kubeadm-config.yaml apiVersion: kubeadm.k8s.io/v1beta1 kind: ClusterConfiguration kubernetesVersion: v1.17.0 controlPlaneEndpoint: "10.21.213.221:6443" apiServer: certSANs: -10.21.213.221networking: podSubnet: 10.244.0.0/16 imageRepository: "registry.aliyuncs.com/google_containers" --- apiVersion: kubeproxy.config.k8s.io/v1alpha1 kind: KubeProxyConfiguration mode: ipvs
检查环境是否已经准备好:
systemctl status dockersystemctl status kubelet
---会报错，但是不用管他，确保已经安装了kubelet就行，后续会自动启动的
初始化
kubeadm init --config /etc/kubernetes/kubeadm-config.yaml
可能遇到的问题：
解决
问题二：
这个报错，可能是因为我分配的内存过小，导致启动很慢，超了40s，所以报错,
解决：
只能重新安装
kubeadm reset --ignore-preflight-errors=Swap rm -rf /var/lib/kubelet/ rm -rf /var/lib/cni/ rm -rf /var/lib/etcd/ rm -rf /etc/kubernetes/ rm -rf /root/.kube/config #不删除，总是报错，可能是因为yaml配置的问题，导致生成镜像无法重复使用 docker rm `docker ps | awk '{print$1}'` docker rm `docker ps -a| awk '{print$1}'`
mkdir /etc/kubernetes systemctl restart kubelet vim /etc/kubernetes/kubeadm-config.yaml kubeadm init --config /etc/kubernetes/kubeadm-config.yaml
成功标识：
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 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/ You can now join any number of control-plane nodes by copying certificate authorities and service account keys on each node and then running the following as root: kubeadm join 10.21.213.221:6443 --token raduip.by90yi3u3crjd2qp \ --discovery-token-ca-cert-hash sha256:05d7114b1b32d67e40d01d63b8f9a72cb9c290636ab198fdb20cb93cf05b6611 \ --control-plane Then you can join any number of worker nodes by running the following on each as root: kubeadm join 10.21.213.221:6443 --token raduip.by90yi3u3crjd2qp \ --discovery-token-ca-cert-hash sha256:05d7114b1b32d67e40d01d63b8f9a72cb9c290636ab198fdb20cb93cf05b6611
按照提示继续操作
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config
iptables -I INPUT -p tcp --dport=6443 -j ACCEPT
curl -fsSL https://docs.projectcalico.org/v3.9/manifests/calico.yaml | sed "s@192.168.0.0/16@10.244.0.0/16@g" | kubectl apply -f -
查看有没有问题
# 查看kubelet的日志
journalctl -fu kubelet
work节点加入master
在222节点上执行
kubeadm join 10.21.213.221:6443 --token raduip.by90yi3u3crjd2qp \ --discovery-token-ca-cert-hash sha256:05d7114b1b32d67e40d01d63b8f9a72cb9c290636ab198fdb20cb93cf05b6611
在master（221）上查看状态
kubectl get nodes NAME STATUS ROLES AGE VERSION 10.21.213.221 Ready master 23m v1.17.0 10.21.213.223 NotReady <none> 2m58s v1.17.0
---一般过了2-5分钟，才会变成Ready
---加入有问题
如果太久了也没有变成ready，那就是有问题，你可以通过命令查看
# kubectl get pods -n kube-system NAME READY STATUS RESTARTS AGE coredns-9d85f5447-8b24d 0/1 ContainerCreating 0 26m coredns-9d85f5447-mp2w7 0/1 ContainerCreating 0 26m etcd-10.21.213.221 1/1 Running 0 26m kube-apiserver-10.21.213.221 1/1 Running 0 26m kube-controller-manager-10.21.213.221 1/1 Running 0 26m kube-proxy-qc2mk 1/1 Running 0 26m kube-proxy-t6bqj 1/1 Running 0 21m kube-proxy-wtpmb 0/1ImagePullBackOff0 6m26s kube-scheduler-10.21.213.221 1/1 Running 0 26m
查看详细内容
# kubectl describe pod kube-proxy-wtpmb -n kube-system Name: kube-proxy-wtpmb Namespace: kube-system Priority: 2000001000 Priority Class Name: system-node-critical Node: 10.21.213.223/10.21.213.223 Start Time: Mon, 02 Mar 2020 15:45:47 +0800 Labels: controller-revision-hash=bdc8bbd47 。。。 Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 8m41s default-scheduler Successfully assigned kube-system/kube-proxy-wtpmb to 10.21.213.223 Warning Failed 7m16s kubelet, 10.21.213.223 Failed to pull image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0": rpc error: code = Unknown desc = dial tcp: lookup dockerauth.cn-hangzhou.aliyuncs.com on [::1]:53: read udp [::1]:33942->[::1]:53: read: connection refused Warning Failed 7m4s kubelet, 10.21.213.223 Failed to pull image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0": rpc error: code = Unknown desc = Error response from daemon: Get https://registry.aliyuncs.com/v2/: dial tcp: lookup registry.aliyuncs.com on [::1]:53: read udp [::1]:49098->[::1]:53: read: connection refused Warning Failed 6m36s kubelet, 10.21.213.223 Failed to pull image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0": rpc error: code = Unknown desc = Error response from daemon: Get https://registry.aliyuncs.com/v2/: dial tcp: lookup registry.aliyuncs.com on [::1]:53: read udp [::1]:55991->[::1]:53: read: connection refused Warning Failed 5m48s (x4 over 7m16s) kubelet, 10.21.213.223Error: ErrImagePull Normal Pulling 5m48s (x4 over 8m17s) kubelet, 10.21.213.223 Pulling image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0" Warning Failed 5m48s kubelet, 10.21.213.223 Failed to pull image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0": rpc error: code = Unknown desc = Error response from daemon: Get https://registry.aliyuncs.com/v2/: dial tcp: lookup registry.aliyuncs.com on [::1]:53: read udp [::1]:41242->[::1]:53: read: connection refused Warning Failed 5m23s (x7 over 7m15s) kubelet, 10.21.213.223Error: ImagePullBackOff Normal BackOff 3m8s (x16 over 7m15s) kubelet, 10.21.213.223 Back-off pulling image "registry.aliyuncs.com/google_containers/kube-proxy:v1.17.0"
上面的报错是说防火墙的问题
不想排错，就直接iptables -F
下面的是如果重新加入node
在master节点上执行
#kubectl get node NAME STATUS ROLES AGE VERSION master.hanli.com Ready master 3d7h v1.13.0 slave1.hanli.com Ready <none> 3d7h v1.13.0 slave2.hanli.com Ready <none> 3d7h v1.13.0 slave3.hanli.com Ready <none> 3d7h v1.13.0 #查看下pod情况 [root@master] ~$ kubectl get pods -o wide NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES curl-66959f6557-r4crd 1/1 Running 1 6m32s 10.244.2.7 slave2.hanli.com <none> <none> nginx-58db6fdb58-5wt7p 1/1 Running 0 3d6h 10.244.1.4 slave1.hanli.com <none> <none> nginx-58db6fdb58-7qkfn 1/1 Running 0 3d6h 10.244.3.2 slave3.hanli.com <none> <none> #封锁node，排干node上的pod #kubectl drain slave3.hanli.com --delete-local-data --force --ignore-daemonsets node/slave3.hanli.com cordoned WARNING: Ignoring DaemonSet-managed pods: kube-flannel-ds-amd64-8hhsb, kube-proxy-6vjcb; Deleting pods with local storage: monitoring-grafana-8445c4b56d-j2wfl pod/monitoring-grafana-8445c4b56d-j2wfl evicted pod/nginx-58db6fdb58-7qkfn evicted node/slave3.hanli.com evicted #删除slave3节点 [root@master] ~$ kubectl delete node slave3.hanli.com node "slave3.hanli.com" deleted
在node接点上执行
在slave3上执行： kubeadm reset ifconfig cni0 down ip link delete cni0 ifconfig flannel.1 down ip link delete flannel.1 rm -rf /var/lib/cni/ 2、重新使node加入集群 使节点加入集群的命令格式是kubeadm join --token<token><master-ip>:<master-port> --discovery-token-ca-cert-hash sha256:<hash>
如果我们忘记了Master节点的token和hash，可以使用下面的命令来查看：
kubeadm token list TOKEN TTL EXPIRES USAGES DESCRIPTION EXTRA GROUPS 默认情况下，token的有效期是24小时，如果token已经过期的话，可以使用以下命令重新生成： [root@master] ~$ kubeadm token create sek6z6.knv9grhe9ggvtts0 如果你找不到–discovery-token-ca-cert-hash的值，可以使用以下命令生成： [root@master] ~$ openssl x509 -pubkey -in /etc/kubernetes/pki/ca.crt | openssl rsa -pubin -outform der 2>/dev/null | openssl dgst -sha256 -hex | sed 's/^.* //' 7845e6615fcae889eedd6fe55174d904ddd4d3ca5257f04c4438cc67cf06ba58 现在登录到工作节点服务器，然后用root权限运行如下命令加入集群 [root@slave3] /var/lib/cni$ kubeadm join 192.168.255.130:6443 --token sek6z6.knv9grhe9ggvtts0 --discovery-token-ca-cert-hash sha256:7845e6615fcae889eedd6fe55174d904ddd4d3ca5257f04c4438cc67cf06ba58 # 稍等即可看到节点已加入 [root@master] ~$ kubectl get nodes NAME STATUS ROLES AGE VERSION master.hanli.com Ready master 3d10h v1.13.2 slave1.hanli.com Ready <none> 3d10h v1.13.2 slave2.hanli.com Ready <none> 3d10h v1.13.2 slave3.hanli.com Ready <none> 85s v1.13.2 原文：https://blog.csdn.net/fanren224/article/details/86610799
Comments | NOTHING