Hey everyone! I'm new here and tackling setting up Kubernetes locally for the first time using kubeadm. I have a control plane node with 16GB RAM and an Intel i7-7700K CPU and a worker node that's got an Intel Xeon E5-2680 v3 CPU with 16GB RAM as well. I'm running into a problem where after a few minutes, my cluster stops responding to calls to the kube API server and gives me a 'connection refused' error. Also, the pods in the kube-system namespace seem to be stuck in a restarting loop. For instance, the coredns pods are still in 'ContainerCreating' status and the kube-apiserver pod isn't fully up. I followed the official guide closely but can't seem to figure out what's going wrong. I've been analyzing the logs with journalctl -xeu kubelet, and I'm seeing issues with pod restarts and communication errors with the kube API server. Any tips on what I might be missing would be greatly appreciated!
2 Answers
Can you share your .yaml configuration file and also the output from 'ip route show' for both nodes? That might help in diagnosing the issue more clearly.
It sounds like there might be some networking issues going on. Have you checked to see how your routing is set up? Sometimes, if the networking isn't configured correctly, it can cause problems with pod connectivity. Make sure that your nodes can communicate with each other properly.
I haven't messed with any network settings yet. I'm not sure where to start with that.
I just updated my post with the routing info. Hopefully, that helps!