k8s coredns pods 由于 属性 max_concurrent 显示 CrashLoopBackOff

Question

我相信这个问题是在 k8s 从 1.18 升级到 1.19 后出现的。升级集群后没检查

k8scka@master:~$ kubectl get pods -n kube-system
NAME                                       READY   STATUS             RESTARTS   AGE
calico-kube-controllers-7dbc97f587-dfbwl   1/1     Running            0          8d
calico-node-jkgwv                          1/1     Running            0          10d
calico-node-wkncc                          1/1     Running            0          10d
coredns-66bff467f8-frh49                   0/1     CrashLoopBackOff   2093       7d10h
coredns-66bff467f8-wlb22                   0/1     CrashLoopBackOff   2092       7d10h
etcd-master                                1/1     Running            0          8d
kube-apiserver-master                      1/1     Running            0          8d
kube-controller-manager-master             1/1     Running            0          8d
kube-proxy-ljz55                           1/1     Running            0          8d
kube-proxy-w8nvg                           1/1     Running            0          8d
kube-scheduler-master                      1/1     Running            0          8d
k8scka@master:~$

Answer 1

验证了 coredns CM 并删除了 属性 max_concurrent。 问题自动解决。

Answer 2

否则， 我们可以升级coredns来解决问题。