liveBook · Manning

Control plane node

As explained in Google Anthos in Action

A control plane is the Kubernetes cluster layer that stores cluster state, accepts requests, schedules workloads, and runs controllers.

A control plane node belongs to Kubernetes’ management layer. It stores cluster state, accepts API requests, schedules workloads, and runs controllers; worker nodes execute the scheduled containers. section 3.2 section 3.5

The control plane is assembled from components with distinct operational responsibilities:

  • etcd stores every Kubernetes resource and the complete cluster state. Run at least three replicas, use an odd count for quorum, back up outside the cluster, and protect the database because it can contain sensitive data such as secrets. section 3.2.2
  • The Kubernetes API server receives cluster requests and exposes the management interface used by other components. section 3.2.1
  • The scheduler assigns workloads to suitable worker nodes, while the controller manager reconciles node status, replication, service endpoints, and service accounts or tokens. section 3.2.1 section 3.2.2
  • The cloud controller manager integrates cloud-specific node resources, routes, load balancers, network filtering, and IP addresses. section 3.2.2

Managed and on-premises operation

In GKE on Google Cloud, Google creates the control plane in a Google-managed project, limiting direct access to administrative nodes and Kubernetes components. In an on-premises GKE installation, operators can inspect control plane resources for troubleshooting and configuration changes. The generated kubeconfig files include targets for the load-balanced admin cluster and an internal debug endpoint. section 3.2

Figure 3.3 Control plane components
03-03

Use the component diagram to trace how etcd, the API server, scheduler, and controllers form the control plane. section 3.2.1

How Google Anthos in Action covers it

How deep it goes
Explains component responsibilities, quorum, backups, security, and deployment access, while deferring detailed etcd procedures.
Examples
Uses GKE console node details, on-premises kubeconfig files, controller role tables, and a control plane component diagram.
What it assumes
Assumes familiarity with Kubernetes clusters and workloads while identifying control plane components and their interactions.
What you can do after
Diagnose control plane problems and evaluate availability, backup, security, and managed-versus-on-premises deployment implications.
How it connects
Connects etcd, Kubernetes API server, scheduler, controller manager, cloud controller manager, worker nodes, and Istio.

Also explained in

Compare how 3 books explain control plane node →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Google Anthos in Action. The section links point to the parts of the book it draws on.