Kubernetes 集群安装 cilium 后 coredns pod 无法启动

昨天在阿里云国内云服务器上用 kubernetes 1.37.1 搭建集群，网络插件用的是 cilium，通过下面的命令安装

cilium install --version 1.20.1 \
  --set ipam.mode=kubernetes \
  --namespace kube-system \
  --set bpf.masquerade=true \
  --set kubeProxyReplacement=true

安装后 cilium 正常运行，但 coredns pod 因健康检查失败无法启动

  spec.containers{coredns}: Liveness probe failed: Get "http://192.168.0.13:8080/health": dial tcp 192.168.0.13:8080: connect: connection refused
  spec.containers{coredns}: Readiness probe failed: Get "http://192.168.0.13:8181/ready": dial tcp 192.168.0.13:8181: connect: connection refused
  spec.containers{coredns}: Container image "registry-vpc.cn-hangzhou.aliyuncs.com/cnblogs-k8s/coredns:v1.14.6" already present on machine and can be accessed by the pod
  spec.containers{coredns}: Container created
  spec.containers{coredns}: Container coredns failed liveness probe, will be restarted
  spec.containers{coredns}: Container started

请问如何解决这个问题？