KubernetesAPIs

Kubernetes APIs are the RESTful interfaces exposed by the Kubernetes control plane for creating, reading, updating, deleting, and watching cluster state. The kube-apiserver serves as the central API server, validating requests, coordinating operations, and persisting resources in etcd. Clients and components interact with the API to manage workloads, services, configurations, and other cluster objects.

The API surface is organized into API groups and versions. The core group is v1, and additional groups include apps/v1, batch/v1, networking.k8s.io/v1, and rbac.authorization.k8s.io/v1, among others. Resources are defined within these groups and may be namespaced or cluster-scoped. Each resource has a defined schema, with common fields such as metadata, spec, and status, and is described by OpenAPI.

Clients perform standard HTTP operations to manipulate resources. Typical methods include GET to read or list, POST to create, PUT or PATCH to modify, and DELETE to remove. The API supports listing and watching resources to receive changes in real time. Data formats are JSON or YAML, and client libraries (such as client-go) or command-line tools like kubectl interact with the API server.

Security and policy are integral to the API. Requests are subject to authentication and authorization (often RBAC), and admission controllers enforce validation, defaulting, and other policy decisions before persistence. Custom resources can be added via CustomResourceDefinitions (CRDs) under the apiextensions.k8s.io/v1 group, enabling users to extend the API with new object types that integrate with the same API machinery.

API versions follow a deprecation policy, and the GA-stable versions are generally safe for production use. The Kubernetes API is designed to be stable, extensible, and interoperable across components and tooling.