Ran into this problem again so I spent more time trying to solve it. Although I haven’t been able to pinpoint the root cause yet, I suspect it’s a containerd issue.
I can reproduce it with following steps:
- kubeadm reset
- sudo ip link set cni0 down && sudo ip link delete cni0
- sudo rm -rf /etc/cni/net.d/
- initialize the k8s again with flannel CNI plugin
restart containerd
can solve this issue
sudo systemctl restart containerd