首先是按照二进制方式安装的k8s。

[root@ht22 calico]# cat /etc/redhat-release
   CentOS Linux release 7.9.2009 (Core)

[root@ht22 calico]# uname -r
   3.10.0-1160.45.1.el7.x86_64

[root@ht22 calico]# docker -v
   Docker version 18.03.0-ce, build 0520e24

在节点上执行命令 kubectl get pod, 出现错误

[root@ht22 network-scripts]# kubectl get pod
The connection to the server localhost:8080 was refused - did you specify the right host or port?

[root@ht22 network-scripts]#   vi  /etc/kubernetes/conf/bootstrap.kubeconfig
     server: https://127.0.0.1:6443  改成master地址 : 10.128.1.156
    [root@ht22 network-scripts]#  rm -rf kubelet.kubeconfig
    [root@ht22 network-scripts]#  systemctl restart kubelet   //重新生成  kubelet.kubeconfig

[root@ht22 calico]# echo "export KUBECONFIG=/etc/kubernetes/conf/kubelet.kubeconfig" >> /etc/profile
   [root@ht22 calico]# source /etc/profile
    
修复之后出现结果：

[root@ht22 calico]# kubectl get nodes
  NAME STATUS ROLES AGE VERSION
  ht10.node NotReady,SchedulingDisabled k8s-node 82d v1.10.13
  ht11.node Ready,SchedulingDisabled k8s-node 79d v1.10.13
  ht12.node Ready,SchedulingDisabled k8s-node 79d v1.10.13
  ht13.node Ready,SchedulingDisabled k8s-node 80d v1.10.13
  ht2.node Ready,SchedulingDisabled k8s-node 99d v1.10.13
  ht20.node Ready,SchedulingDisabled k8s-node 29d v1.10.13
  ht22.node Ready,SchedulingDisabled k8s-node 23h v1.10.13