k8s集群添加master节点报control plane 错误

本文分享自天翼云开发者社区《k8s集群添加master节点报control plane 错误》，作者：SummerSnow
背景介绍

1. 在刚部署的k8s集群中添加新的master节点时，报了error execution phase preflight: One or more conditions for hosting a new control plane instance is not satisfied的错误，接下来就针对此问题进行解决说明。
2. 环境介绍

复制代码

 
k8s版本

1. kubectl version
2. Client Version: version.Info{Major:"1", Minor:"23", GitVersion:"v1.23.5", GitCommit:"xxx", GitTreeState:"clean", BuildDate:"2022-03-16T15:58:47Z", GoVersion:"go1.17.8", Compiler:"gc", Platform:"linux/amd64"}

复制代码

 
部署方式

1. kubeadm

复制代码

当前节点信息

1. kubectl get node
2. NAME          STATUS   ROLES                  AGE     VERSION
3. k8s-master1   Ready    control-plane,master   24h     v1.23.5

复制代码

master节点加入

1. #第1步 打印加入集群的相关信息
2. kubeadm token create --print-join-command
3. kubeadm join xx.xx.xx.xx:6443 --token xxxx.xxxx --discovery-token-ca-cert-hash sha256:xxx
4. ​
5. #第2步 打印加入master的certs信息
6. kubeadm init phase upload-certs --upload-certs
7. I0520 14:51:22.848075    1096 version.go:255] remote version is much newer: v1.30.1; falling back to: stable-1.23
8. [upload-certs] Storing the certificates in Secret "kubeadm-certs" in the "kube-system" Namespace
9. [upload-certs] Using certificate key:
10. aaaaaxxxxxxxxxxxxxxxxxx
11. ​
12. #第3步 信息拼接（第一步和第二步）
13. kubeadm token create --print-join-command
14. kubeadm join xx.xx.xx.xx:6443 --token xxxx.xxxx --discovery-token-ca-cert-hash sha256:xxx --control-plane --certificate-key aaaaaxxxxxxxxxxxxxxxxxx
15. ​
16. #第4步 在待加入节点上执行上面第3步的命令，然后报错信息如下：
17. [preflight] Running pre-flight checks
18.      [WARNING Hostname]: hostname "xxx" could not be reached
19.      [WARNING Hostname]: hostname "xxx": lookup xxx on 114.114.114.114:53: no such host
20.      [WARNING Service-Kubelet]: kubelet service is not enabled, please run 'systemctl enable kubelet.service'
21. [preflight] Reading configuration from the cluster...
22. [preflight] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -o yaml'
23. error execution phase preflight:
24. One or more conditions for hosting a new control plane instance is not satisfied.
25. ​
26. unable to add a new control plane instance to a cluster that doesn't have a stable controlPlaneEndpoint address
27. ​
28. Please ensure that:
29. * The cluster has a stable controlPlaneEndpoint address.
30. * The certificates that must be shared among control plane instances are provided.
31. ​
32. ​
33. To see the stack trace of this error execute with --v=5 or higher

复制代码

解决方法

方法1

1. #1 查看 kubeadm-config的信息(只截取了一部分信息)
2. kubectl get  cm kubeadm-config -n kube-system -oyaml
3. ​
4. apiVersion: v1
5. data:
6.    ClusterConfiguration: |
7.      apiServer:
8.        extraArgs:
9.          authorization-mode: Node,RBAC
10.        timeoutForControlPlane: 4m0s
11.      apiVersion: kubeadm.k8s.io/v1beta3
12.      certificatesDir: /etc/kubernetes/pki
13.      clusterName: kubernetes
14.      controllerManager: {}
15.      dns: {}
16.      etcd:
17.        local:
18.          dataDir: /var/lib/etcd
19.      imageRepository: xxx
20.      kind: ClusterConfiguration
21.      kubernetesVersion: v1.23.5
22.      networking:
23.        dnsDomain: cluster.local
24.        podSubnet: xx.xx.0.0/16
25.        serviceSubnet: xx.xx.0.0/16
26.      scheduler: {}
27. ​
28. #2 添加controlPlaneEndpoint
29. kubectl edit cm kubeadm-config -n kube-system
30. ​
31. apiVersion: v1
32. data:
33.    ClusterConfiguration: |
34.      apiServer:
35.        extraArgs:
36.          authorization-mode: Node,RBAC
37.        timeoutForControlPlane: 4m0s
38.      apiVersion: kubeadm.k8s.io/v1beta3
39.      certificatesDir: /etc/kubernetes/pki
40.      clusterName: kubernetes
41.      controllerManager: {}
42.      dns: {}
43.      etcd:
44.        local:
45.          dataDir: /var/lib/etcd
46.      imageRepository: xxx
47.      kind: ClusterConfiguration
48.      kubernetesVersion: v1.23.5
49.      #在下面添加一行信息
50.      controlPlaneEndpoint: "xx.xx.xx.xx:port" ## 换成自己的ip和port
51.      networking:
52.        dnsDomain: cluster.local
53.        podSubnet: xx.xx.0.0/16
54.        serviceSubnet: xx.xx.0.0/16
55.      scheduler: {}

复制代码

方法2

如果k8s集群中只有一个节点，可将k8s集群重置，然后重新初始化

1. #1 重置k8s，然后删除相关文件
2. kubeadm reset -f
3. rm -rf /etc/kubernetes
4. rm -rf ~/.kube
5. ​
6. #2 重新初始化
7. kubeadm init --kubernetes-version 1.23.5 --control-plane-endpoint "xx.xx.xx.xx:port" --pod-network-cidr=xx.xx.xx.xx/16 --service-cidr=xx.xx.xx.xx/16 --upload-certs
8.   

复制代码

 

来源：程序园用户自行投稿发布，如果侵权，请联系站长删除
免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！