项目场景：

使用kubeadm搭建k8s集群

环境：CentOS7.9

---

问题描述

部署flannel网络插件失败：

NAMESPACE      NAME                             READY   STATUS              RESTARTS   AGE
kube-flannel   kube-flannel-ds-55nbz            0/1     CrashLoopBackOff    3          2m48s

---

原因分析：

查看pod日志：kubectl get pod -n kube-flannel kube-flannel-ds-55nbz

E0825 09:00:25.015344       1 main.go:330] Error registering network: failed to acquire lease: subnet "10.244.0.0/16" specified in the flannel net config doesn't contain "172.16.0.0/24" PodCIDR of the "master" node.
W0825 09:00:25.018642       1 reflector.go:436] github.com/flannel-io/flannel/subnet/kube/kube.go:403: watch of *v1.Node ended with: an error on the server ("unable to decode an event from the watch stream: context canceled") has prevented the request from succeeding
I0825 09:00:25.018698       1 main.go:447] Stopping shutdownHandler...

原因：初始化master节点时，指定了–pod-network-cidr=172.16.0.0/16
那么，kube-flannel.yml中该部分网段也要进行对应，默认是“10.244.0.0/16”

---

解决方案：

vi kube-flannel.yml

 
 net-conf.json: |
    {
      "Network": "172.16.0.0/16",
      "Backend": {
        "Type": "vxlan"
      }
    }

kubectl apply -f kube-flannel.yml

再对pod进行观察

watch -n1 kubectl get pods -A