master中“kubectl get node ”STATUS显示notready

错误显示：

kubectl get node

kubectl describe node 10.0.3.104

Conditions:
  Type             Status    LastHeartbeatTime                 LastTransitionTime                Reason                    Message
  ----             ------    -----------------                 ------------------                ------                    -------
  OutOfDisk        Unknown   Tue, 29 Jan 2019 11:06:05 +0800   Tue, 29 Jan 2019 11:06:45 +0800   NodeStatusUnknown         Kubelet stopped posting node status.
  MemoryPressure   Unknown   Tue, 29 Jan 2019 11:06:05 +0800   Tue, 29 Jan 2019 11:06:45 +0800   NodeStatusUnknown         Kubelet stopped posting node status.
  DiskPressure     Unknown   Tue, 29 Jan 2019 11:06:05 +0800   Tue, 29 Jan 2019 11:06:45 +0800   NodeStatusUnknown         Kubelet stopped posting node status.
  PIDPressure      False     Tue, 29 Jan 2019 11:06:05 +0800   Mon, 28 Jan 2019 15:11:01 +0800   KubeletHasSufficientPID   kubelet has sufficient PID available
  Ready            Unknown   Tue, 29 Jan 2019 11:06:05 +0800   Tue, 29 Jan 2019 11:06:45 +0800   NodeStatusUnknown         Kubelet stopped posting node status.

显示：Kubelet stopped posting node status.
继而在node104节点中，查看

tail /var/log/messages

解决办法：

<1> 重启node中的各个组件

systemctl daemon-reload
systemctl restart docker
systemctl restart kubelet
systemctl restart kube-proxy

且通过ps -ef |grep kube查看各个组件的运行情况，注意看是否是当前时间的重启情况。如果未启动kubelet，在master中也是检查不到node的。

<2> 查看防火墙问题

systemctl status firewalld
----------执行
systemctl stop firewalld
systemctl enable firewalld

确保防火墙关闭

在LB负载均衡的服务器上，查看运行日志，可以实时监测k8s运行部署情况：