systemd is the Linux service manager relevant to Kubernetes node administration. Unlike kube-proxy and CNI, which run as DaemonSets, kubelet runs directly as a Linux background service on the node. section 1.8
Use systemd tools to inspect and operate kubelet:
- List service unit files and locate kubelet: `sudo systemctl list-unit-files --type service | grep kubelet -A3 -B3`. section 1.8
- Start a stopped kubelet with `systemctl start kubelet`. section 1.8
- Enable kubelet with `systemctl enable kubelet` so it starts after node reboot. section 1.8
- Check whether kubelet is active with `systemctl status kubelet`. section 1.8
- Use `journalctl` to inspect logs collected by systemd-journald when diagnosing kubelet or node-level failures. section 1.8
Repairing a node service
If a node problem suggests that kubelet is down, list its unit entry, check it with `systemctl status kubelet`, start it if stopped, and inspect its journal when the failure persists. Confirm the resulting Kubernetes state afterward. section 1.8
$ systemctl list-unit-files --type service | grep kubelet -A3 -B3 iscsid.service disabled enabled kmod-static-nodes.service static - kmod.service alias - kubelet.service enabled enabled modprobe@.service static - nfs-common.service masked enabled nfs-idmapd.service static -
The service listing makes kubelet’s enabled state visible alongside other Linux units, providing a quick node-level inspection reference. section 1.8
