Static Pods
K8s provides a way to “override” the scheduler and deploy pods statically. These static pods are managed directly by the kubelet daemon on a specific node, without the API server observing them.
The kubelet automatically tries to create a mirror Pod on the Kubernetes API server for each static Pod. This means that the Pods running on a node are visible on the API server, but cannot be controlled from there.
The static pod names will be suffixed with the node hostname with a leading hyphen.
Noticeably, static pod deployments are used by the kubeadm tool to deploy control plane components (e.g api server, scheduler, etc.)
Deploying static pods
The kubelet read static pod definitions for a defined folder. This can either be passed as an argument --pod-manifest-path or as the staticPodPath part of the configuration file (--config) to the kubelet.service.
systemctl cat kubelet.service can be useful to find the path. Most of the times is /etc/kubernetes/manifests
To deploy a static pod, the only required action is to create a pod definition in this folder. The kubelet will pick it up automatically and deploy it. Similarly, for any changes done to the file, the kubelet will reflect them in the deployment.