Installing charts in Kubernetes - Step-by-Step CLI Walkthrough

Introduction

Installing charts helps you add ready-made applications to your Kubernetes cluster easily. Charts package all the resources needed to run an app, so you don't have to create each part manually.

When you want to deploy a database like MySQL quickly without writing all Kubernetes files.

When you need to add monitoring tools like Prometheus to your cluster with minimal setup.

When you want to install a web server such as Nginx with default settings fast.

When you want to try out an application without configuring every detail yourself.

When you want to manage app upgrades and rollbacks easily using a package system.

Commands

This command adds the Bitnami chart repository to Helm so you can access many ready-to-use charts from Bitnami.

Terminal

helm repo add bitnami https://charts.bitnami.com/bitnami

Expected Output

"bitnami" has been added to your repositories

This updates your local list of charts from all added repositories to make sure you have the latest versions.

Terminal

helm repo update

Expected Output

Hang tight while we grab the latest from your chart repositories...

...Successfully got an update from the "bitnami" chart repository

Update Complete.

This installs the Nginx chart from Bitnami repository into your Kubernetes cluster with the release name 'my-nginx'.

Terminal

helm install my-nginx bitnami/nginx

Expected Output

NAME: my-nginx

LAST DEPLOYED: Sat Jun 10 12:00:00 2023

NAMESPACE: default

STATUS: deployed

REVISION: 1

NOTES:

1. Get the application URL by running these commands:

  export POD_NAME=$(kubectl get pods --namespace default -l app.kubernetes.io/name=nginx -o jsonpath="{.items[0].metadata.name}")

  echo "Visit http://127.0.0.1:8080 to use your application"

  kubectl port-forward $POD_NAME 8080:80

→

--namespace - Specify the Kubernetes namespace to install the chart into

→

--set - Override chart default values during installation

This command checks the status of pods in your cluster to verify that the Nginx pod is running after installation.

Terminal

kubectl get pods

Expected Output

NAME                             READY   STATUS    RESTARTS   AGE

my-nginx-nginx-5d7f9d7f7b-abcde 1/1     Running   0          30s

Key Concept

If you remember nothing else from this pattern, remember: Helm charts let you install complex apps on Kubernetes with one simple command.

Common Mistakes

Trying to install a chart without adding the repository first

Helm won't find the chart because it doesn't know where to look for it.

Always add the chart repository using 'helm repo add' before installing charts from it.

Not running 'helm repo update' after adding a new repository

Your local list of charts will be outdated, so you might not get the latest chart versions.

Run 'helm repo update' to refresh your chart list before installing.

Not checking pod status after installation

You might think the app is running when it actually failed to start.

Use 'kubectl get pods' to confirm the app pods are running properly.

Summary

Add a chart repository with 'helm repo add' to access ready-made apps.

Update your local chart list with 'helm repo update' to get the latest versions.

Install an app using 'helm install' with a release name and chart name.

Verify the app is running by checking pods with 'kubectl get pods'.

Practice

1. What is the primary purpose of using helm install in Kubernetes?

easy

A. To deploy an application packaged as a Helm chart

B. To delete a Kubernetes namespace

C. To update the Kubernetes cluster version

D. To create a new Kubernetes node

Solution

1. Step 1: Understand Helm's role

   Helm packages Kubernetes apps into charts for easy deployment.

2. Step 2: Purpose of helm install

   This command deploys the packaged app (chart) into the cluster.

3. Final Answer:

   To deploy an application packaged as a Helm chart -> Option A

4. Quick Check:

   Helm install = deploy app [OK]

Hint: Helm install means deploy app from chart [OK]

Common Mistakes:

• Confusing install with delete or update commands
• Thinking helm install creates nodes
• Assuming helm install manages cluster versions

2. Which of the following is the correct syntax to install a Helm chart named nginx with release name myweb?

easy

A. helm install nginx myweb

B. helm create myweb nginx

C. helm install myweb nginx

D. helm deploy myweb nginx

Solution

1. Step 1: Recall Helm install syntax

   The correct syntax is helm install [release-name] [chart-name].

2. Step 2: Match syntax to options

   helm install myweb nginx uses helm install myweb nginx, which matches the syntax.

3. Final Answer:

   helm install myweb nginx -> Option C

4. Quick Check:

   helm install release chart = helm install myweb nginx [OK]

Hint: Remember: helm install release-name chart-name [OK]

Common Mistakes:

• Swapping release name and chart name order
• Using helm create or helm deploy instead of install
• Omitting release name

3. What will be the output of the command helm install myapp bitnami/nginx --namespace web if the namespace web does not exist?

medium

A. Helm will prompt to create the namespace interactively

B. Helm will create the namespace web and install the chart

C. Helm will install the chart in the default namespace

D. Installation will fail with an error about missing namespace

Solution

1. Step 1: Understand namespace behavior

   Helm does not create namespaces automatically during install.

2. Step 2: Effect of missing namespace

   If the specified namespace web does not exist, Helm will error out.

3. Final Answer:

   Installation will fail with an error about missing namespace -> Option D

4. Quick Check:

   Missing namespace causes install error [OK]

Hint: Helm won't create namespaces automatically [OK]

Common Mistakes:

• Assuming Helm auto-creates namespaces
• Expecting install to default namespace silently
• Thinking Helm prompts for namespace creation

4. You run helm install myapp nginx --set replicaCount=3 but the app still runs with 1 replica. What is the likely cause?

medium

A. You must specify the namespace with --namespace

B. The chart does not support the replicaCount parameter

C. The --set flag syntax is incorrect

D. The Helm client version is outdated

Solution

1. Step 1: Check parameter support

   Not all charts support all parameters; replicaCount must be defined in the chart's values.

2. Step 2: Effect of unsupported parameter

   If unsupported, setting replicaCount has no effect, so default 1 replica remains.

3. Final Answer:

   The chart does not support the replicaCount parameter -> Option B

4. Quick Check:

   Unsupported param means no effect [OK]

Hint: Check if chart supports your --set keys [OK]

Common Mistakes:

• Assuming all charts accept replicaCount
• Thinking namespace affects replica count
• Believing --set syntax is always wrong
• Blaming Helm version without checking chart

5. You want to install a Helm chart myapp with a custom image tag v2.0 and in namespace production. Which command correctly does this in one step?

hard

A. helm install myapp ./myapp-chart --set image.tag=v2.0 --namespace production

B. helm install ./myapp-chart myapp --set image.tag=v2.0 --namespace production

C. helm install myapp ./myapp-chart --image.tag=v2.0 --namespace production

D. helm install myapp ./myapp-chart --set image=v2.0 --namespace production

Solution

1. Step 1: Confirm Helm install syntax

   The syntax is helm install [release-name] [chart-path] [flags].

2. Step 2: Correct use of --set and namespace

   Use --set image.tag=v2.0 to override image tag and --namespace production to specify namespace.

3. Final Answer:

   helm install myapp ./myapp-chart --set image.tag=v2.0 --namespace production -> Option A

4. Quick Check:

   Release, chart, --set key=value, --namespace [OK]

Hint: Use --set key=value and --namespace together [OK]

Common Mistakes:

• Swapping release name and chart path
• Using --image.tag instead of --set
• Setting wrong key like image=v2.0
• Omitting namespace flag