Kay Yan

> Hi @yankay, I wasn't aware of OOMScoreAdjust, so that value overrides whatever we put in config.toml? Either way, IMO oom_score should be changed in config.toml just for consistency, either...

HI @Hamza-Rana1 , we can follow the docs : https://github.com/kubernetes-sigs/kubespray/blob/master/docs/cilium.md#kube-proxy-replacement-with-cilium

I think there are some problems, about the kube-proxy or CNI.

> I would go with option 3, workarounds like 1 and 2 tend to stick around for too long past their due time. Thank you for suggestion.

HI @enneitex , Thanks for you PR. You can use ``` git commit --amend git push *** --force ``` to trigger the CI again.

HI @floryut , I'm sorry for there is a typo: Fix ETCD memory leak issue by adding max_concurrent=1000 in the CoreDNS config. (https://github.com/kubernetes-sigs/kubespray/pull/9307, @yankay) => Fix CoreDNS memory leak issue...

What's your ansible and python version? I guess we should use the right version https://github.com/kubernetes-sigs/kubespray/blob/master/requirements-2.12.txt .

Maybe the right commands is ``` ansible-playbook -i /inventory/sample/hosts.yml --become --become-user=root --user=ubuntu --private-key /root/.ssh/id_rsa cluster.yml ``` And the `/inventory/sample/hosts.yml` should be right. We can follow the usage method step by...

> @yankay What could be the cause of this error?  I guess there is some download error. It's slow in China. BTW， you can try a faster way with...

HI @nmr. I guess if we want to avoid "ImagePullBackOff ", the better way is to Install a registry in the Air-Gap environment.