Kubelete is always in status of auto-restart. Why?

I am following the method to install k8s and deploy tao-toolkit-api.

However. Before I run the command sudo kubeadm init --pod-network-cidr=192.168.0.0/16, I met the problem that kubelet is always in status of auto-restart.

-The driver of Kubelete and docker is systemd, and I appended the content “exec-opts”: [“native.cgroupdriver=systemd”] to /etc/docker/daemon.json.

-What’s more, I added the content Environment=“KUBELET_SYSTEM_PODS_ARGS=–pod-manifest-path=/etc/kubernetes/manifests --allow-priviledged=true --fail-swap-on=false” into the /etc/systemd/system/kubelet.service.d/10-kubeadm.conf

After I ran the command sudo systemctl status kubelet, it displayed that kubelete didn’t start

How should I do to deal with the problems ?

The picture below is the contents of journalctl -xefu kubelet

The picture below is the contents of journalctl -u kubelet

Thank you for your help is advanced

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

H @swka1043338
Currently let’s focus on the official way mentioned in TAO doc.
We can continue to sync in AutoML training speed and GPU problem - #23 by Morganh

Please upload the log as I requested. I find that you are using P100. This architecture is not mentioned in setup requirement of TAO-API.
So, I suggest you upload the full log in AutoML training speed and GPU problem - #23 by Morganh
Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.