There is a rule to drop packets with ctstate INVALID in the KUBE-FORWARD chain.
Since the communication conditions are not determined, communication other than kube-proxy is also dropped and I am in trouble.
Even if the rule is deleted, it will be restored by the timer, so please improve it so that you can specify whether to set the rule in the config setting.
Cluster information:
Kubernetes version:1.21
Cloud being used: (put bare-metal if not on a public cloud)
Installation method:Microk8s
Host OS: ubuntu20.04
CNI and version:calico v3.19.1
CRI and version: containerd v1.5.2