Install Your First Helm Chart on Kubernetes

Helm is the de-facto package manager for Kubernetes: it plays the same role that apt, brew, or pip play in their ecosystems. Instead of hand-applying raw manifests, you pull a pre-built, configurable bundle called a chart from a repository and install it into your cluster with a single command. Nearly every major CNCF project ships as a Helm chart today, so platform engineers spend far more time consuming charts than authoring them. The consumer skillset covers roughly 90% of real-world Helm interaction.

Under the hood, Helm is a thin client-side renderer. When you run helm install, it templates the chart against your values, applies the resulting plain manifests through the Kubernetes API, and records the install as a release, storing the rendered manifests as a Secret in the release's namespace. That release history is what makes helm history and helm rollback possible.

In this lab, you will take on the consumer half of Helm end to end. You will add the upstream podinfo chart repository, search and inspect the chart's metadata and default values, and install a pinned version into a dedicated namespace. From there, you will override values with a small values.yaml file, upgrade the release with --atomic so a bad change rolls back automatically, walk through the release history, roll back to an earlier revision, and uninstall the release cleanly.

Objectives

By completing this beginner lab, you will be able to:

• Add a public chart repository with helm repo add and inspect charts with helm show chart and helm show values
• Install a pinned version of the podinfo chart into a dedicated namespace and explore the release with helm list and helm status
• Override default chart values via a values.yaml file and apply the change with helm upgrade --atomic
• Inspect the release history with helm history and roll back to a previous revision with helm rollback
• Uninstall a release and contrast --keep-history against a full removal

Who is this lab for?

This lab is designed for:

• Platform and DevOps engineers who know the core Kubernetes objects but have never installed software via Helm
• Application developers preparing for the CKAD exam
• Anyone who has used kubectl apply directly and is ready to graduate to versioned, repeatable installs