Study guide

2.4 Kubernetes Releases, API Versioning & Deprecation Policy

Kubernetes and Cloud Native Associate KCNA

Track your progress as you read

Finished sections get a check mark.

1 Chapter 1: Introduction & Exam Overview

1.1 About the KCNA Exam & Certification Path 1.2 How to Use This Study Guide

2 Chapter 2: Kubernetes Fundamentals - Cluster Architecture & the API

2.1 Kubernetes Architecture & Control Plane Components 2.2 Worker Node Architecture & Node Components 2.3 Kubernetes API, Declarative Management & kubectl 2.4 Kubernetes Releases, API Versioning & Deprecation Policy

3 Chapter 3: Kubernetes Fundamentals - Workloads & Configuration

3.1 Core Workload Objects (Pods, Deployments & ReplicaSets) 3.2 Advanced Controllers (StatefulSets, DaemonSets, Jobs & CronJobs) 3.3 Namespaces, Labels, Selectors & Annotations 3.4 ConfigMaps, Secrets & Application Configuration 3.5 Pod & Container Lifecycle, Restart Policies & Graceful Termination

4 Chapter 4: Kubernetes Fundamentals - Scheduling & Autoscaling

4.1 Scheduling, Resource Management & Node Selection 4.2 Advanced Scheduling: Affinity, Topology Spread, Priority & Disruption Budgets 4.3 Autoscaling Workloads & Clusters (HPA, VPA, Cluster Autoscaler, KEDA)

5 Chapter 5: Kubernetes Fundamentals - Cluster Administration & Extensibility

5.1 Kubernetes Distributions, Installation Tools & Managed Services 5.2 Cluster Administration: Node Maintenance, Upgrades & etcd Backup 5.3 Extending Kubernetes: CRDs, Operators & the Controller Pattern

6 Chapter 6: Kubernetes Fundamentals - Containerization

6.1 Container Fundamentals, Runtimes & Registries 6.2 Container Image Anatomy: Layers, Tags, Digests & Pull Policy

7 Chapter 7: Container Orchestration - Networking & Service Discovery

7.1 Kubernetes Networking, Services & Ingress 7.2 Cluster DNS & Service Discovery with CoreDNS 7.3 NetworkPolicies, Zero-Trust Segmentation & the Gateway API

8 Chapter 8: Container Orchestration - Storage

8.1 Storage Concepts (PVs, PVCs, StorageClasses & CSI) 8.2 Volume Types, Snapshots & Stateful Storage Operations

9 Chapter 9: Container Orchestration - Security

9.1 Kubernetes Security & Access Control (RBAC & ServiceAccounts) 9.2 The 4C's of Cloud Native Security & Supply-Chain Trust 9.3 Service Mesh, mTLS & Zero-Trust Service Identity

10 Chapter 10: Container Orchestration - Troubleshooting

10.1 Container Troubleshooting & Health Probes 10.2 Cluster & Node-Level Troubleshooting

11 Chapter 11: Cloud Native Application Delivery

11.1 Cloud Native CI/CD Pipelines & Build Tooling 11.2 GitOps & Package Management (Helm, Kustomize, Argo CD & Flux) 11.3 Deployment Strategies & Release Management 11.4 Progressive Delivery Automation: Argo Rollouts, Flagger & Feature Flags 11.5 Application Debugging & Developer Workflows in Kubernetes 11.6 Platform Engineering & Developer Self-Service

12 Chapter 12: Cloud Native Architecture - Observability & Principles

12.1 Cloud Native Observability (Logging, Metrics & Tracing) 12.2 Cloud Native Principles & Architecture Patterns 12.3 Serverless, FaaS & Event-Driven Architecture

13 Chapter 13: Cloud Native Architecture - CNCF Ecosystem & Community

13.1 The CNCF Landscape, Governance & Project Lifecycle 13.2 Cloud Native Community, Collaboration & Contribution

2.4 Kubernetes Releases, API Versioning & Deprecation Policy

Key Takeaways Explain This Topic AI

Kubernetes ships roughly three minor releases per year, and each minor release is supported with patch releases for about 14 months (12 months of standard support plus a 2-month maintenance window).

API maturity is encoded in the apiVersion string: v1alpha1 is experimental and disabled by default, v1beta1 is enabled by default but may still change, and v1 is stable with backward-compatibility guarantees.

The Kubernetes deprecation policy requires GA (v1) API elements to be supported for at least 12 months or 3 minor releases after deprecation, whichever is longer.

A Kubernetes version is written MAJOR.MINOR.PATCH (for example 1.34.2); the minor number identifies the feature release and the patch number identifies the bug-fix build.

Removed APIs are the most common cause of a broken cluster upgrade, which is why kubectl surfaces deprecation warnings and why manifests should be migrated before the removal release lands.

Last updated: August 2026

2.4 Kubernetes Releases, API Versioning & Deprecation Policy

Quick Answer: Kubernetes publishes about three minor releases per year , each supported for roughly 14 months of patch releases. Every API object carries an apiVersion whose suffix encodes maturity: alpha (experimental, off by default, may vanish without notice), beta (on by default, schema may still change), and stable/GA ( v1 , backward compatible). The deprecation policy guarantees a stable API element stays available for at least 12 months or 3 minor releases after it is deprecated, whichever is longer.

KCNA is a conceptual exam, but it is written against a moving platform. Almost every Kubernetes question you will meet in the wild — "why did my manifest stop working after the upgrade?", "is this field safe to depend on?" — comes down to release cadence and API maturity. This section gives you the vocabulary.

1. How Kubernetes Versions Are Numbered

Kubernetes follows semantic versioning in the form MAJOR.MINOR.PATCH :

Segment Example Meaning

MAJOR 1 Has never been incremented past 1 since the 1.0 release in July 2015. A major bump would signal a breaking rewrite.

MINOR 34 The feature release . New features, new APIs, graduations, and removals land here. Roughly three per year.

PATCH 2 Bug fixes and CVE fixes only. Never adds features and never removes APIs — always safe to apply.

The project ships minor releases on a published schedule managed by SIG Release , and each release has a named release team, an enhancements freeze, a code freeze, and a formal release-notes document. Because minor releases arrive roughly every four months, a cluster that is two years behind is already outside the supported window.

Support Window

Each minor release receives approximately 12 months of standard patch support plus an additional 2-month maintenance window , for a practical support life of about 14 months . Managed Kubernetes services (EKS, GKE, AKS) build their own, sometimes longer, support commitments on top of this upstream schedule. The operational consequence is simple and heavily tested: you cannot skip upgrades indefinitely . Falling more than three minor versions behind means you can no longer take a supported path forward.

2. Reading an apiVersion String

Every object manifest begins with an apiVersion . That single line tells you both where the resource lives in the API and how much you can trust it . apiVersion: apps/v1 # named group "apps", stable version v1

  • apiVersion: v1 # core group (no group name), stable
  • apiVersion: batch/v1 # named group "batch", stable
  • apiVersion: networking.k8s.io/v1 # named group, stable
  • apiVersion: flowcontrol.apiserver.k8s.io/v1beta3 # beta — may change
  • The Three Maturity Levels

Level Suffix pattern Enabled by default? Stability promise Typical use

Alpha v1alpha1 , v2alpha1 No — must be switched on with a feature gate None. May be removed or redesigned in any release. Data loss is possible. Short-lived experiments, early feedback

Beta v1beta1 , v2beta2 Yes (historically; newer betas are increasingly off by default) The feature will not be dropped, but the schema may change and a conversion may be required Pre-production evaluation

Stable / GA v1 , v2 Yes Backward compatible; covered by the deprecation policy Everything you run in production

Exam trap: "Beta means production ready." It does not. Beta means enabled by default, not frozen . The classic real-world example is the autoscaling API, where autoscaling/v2beta1 and autoscaling/v2beta2 both existed before autoscaling/v2 stabilised, and manifests had to be rewritten twice.

3. The Deprecation Policy

Kubernetes publishes a formal deprecation policy so that operators can plan migrations instead of discovering breakage during an upgrade.

GA ( v1 ) API versions may be marked deprecated, but must not be removed within a major version of Kubernetes . In practice a stable API you depend on today keeps being served for the whole of the 1.x line.

Beta API versions are deprecated no more than 9 months or 3 minor releases after introduction (whichever is longer), and stop being served 9 months or 3 minor releases after deprecation (whichever is longer).

Alpha API versions may be removed in any release, without prior deprecation notice .

A deprecated element keeps working exactly as before during the deprecation window — deprecation is an announcement, not a behaviour change.

How Deprecations Surface

Release notes carry a dedicated "Deprecations and Removals" section for every minor release.

kubectl prints a warning when you submit an object using a deprecated API version — the API server returns the notice in a Warning HTTP header.

The apiserver_requested_deprecated_apis metric lets Prometheus alert on deprecated API usage before the upgrade, which is the professional way to plan a migration.

The Canonical Failure

The most-cited example is the removal of the old extensions/v1beta1 Ingress and Deployment endpoints in Kubernetes v1.22 , and the removal of PodSecurityPolicy in v1.25 . In both cases the objects simply stopped being accepted, so clusters upgraded without a manifest migration lost the ability to create those resources. The lesson KCNA wants: removals happen at minor releases, so an upgrade is a migration event, not a maintenance event.

4. Feature Gates

A feature gate is a named boolean flag passed to control plane components and the kubelet that turns an in-development feature on or off, for example --feature-gates=InPlacePodVerticalScaling=true . Feature gates are how alpha features are shipped without exposing them to everyone. As a feature graduates, its gate flips to on-by-default at beta and is eventually removed entirely once the feature is GA and unconditional.

5. Version Skew: What Is Allowed to Differ

Kubernetes only supports a bounded difference between component versions, which is why upgrades follow a strict order.

  • Component Allowed skew relative to kube-apiserver
  • kube-apiserver (other instances in an HA set) At most 1 minor version older
  • kube-controller-manager, kube-scheduler, cloud-controller-manager At most 1 minor version older
  • kubelet Up to 3 minor versions older (never newer)
  • kube-proxy Matches its node's kubelet
  • kubectl Within 1 minor version either side

The practical rule: upgrade the control plane first, then the nodes, one minor version at a time. Jumping 1.28 straight to 1.31 is unsupported.