説明
This is a Bug Report
Problem: I was facing an issue, that flannel CNI was throwing errors. A lot of people were facing to the same issue and fix was:
sudo modprobe overlay
sudo modprobe br_netfilter
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
Proposed Solution: Copy paste from en doc to jp.
Page to Update: https://kubernetes.io/docs/setup/production-environment/container-runtimes/#prerequisite-ipv4-forwarding-optional https://kubernetes.io/ja/docs/setup/production-environment/container-runtimes/#ipv4%E3%83%95%E3%82%A9%E3%83%AF%E3%83%BC%E3%83%87%E3%82%A3%E3%83%B3%E3%82%B0%E3%82%92%E6%9C%89%E5%8A%B9%E5%8C%96%E3%81%97-iptables%E3%81%8B%E3%82%89%E3%83%96%E3%83%AA%E3%83%83%E3%82%B8%E3%81%95%E3%82%8C%E3%81%9F%E3%83%88%E3%83%A9%E3%83%95%E3%82%A3%E3%83%83%E3%82%AF%E3%82%92%E8%A6%8B%E3%81%88%E3%82%8B%E3%82%88%E3%81%86%E3%81%AB%E3%81%99%E3%82%8B
based on uri, i hope it should be the same heading :)
Latest version: 1.33
I see that pre-flight check were removed: https://github.com/kubernetes/kubernetes/pull/123464 And the problem might be more "CNI problem", than kubernetes problem, but atleast i want to point out of the inconsistency in different language mutations.
Discusses:
- https://discuss.kubernetes.io/t/pod-running-from-the-same-worker-node-can-not-communicate-each-other-via-service/30774/9
- https://github.com/siderolabs/talos/issues/10430
- https://github.com/kubernetes/kubernetes/issues/122174
Might be two different problem but with the same solution, which is in JP doc but not in EN.