Skip to content

Installing Charts

Now that we got the Helm CLI installed, we can go ahead and install our first chart. A chart is a collection of files that describe a related set of Kubernetes resources. A single chart might be used to deploy something simple, like a single pod, or something complex, like a full web app stack with HTTP servers, databases, caches, and so on.

Configuring the Repository

Similar to many OS package managers, Helm is able to fetch charts from remote repositories, compressed as .tar.gz archives. We can add the podinfo Helm chart to our local Helm installation:

helm repo add podinfo https://stefanprodan.github.io/podinfo

Once the repository is added, we can search for charts in it:

helm search repo podinfo

We can see that the repository contains a single chart, podinfo/podinfo, in Chart version 6.4.1 and App version 6.4.1. The Chart version is the version of the chart itself, while the App version is the version of the application that the chart installs.

NAME                    CHART VERSION   APP VERSION DESCRIPTION
examples/hello-world    0.1.0           1.16.0      A Helm chart for Kubernetes

Configuring the Chart

As mentioned before, charts can be configured using values. Values are a set of parameters ranging from numbers to strings to complex structures such as lists or nested objects. We can see the default values for the podinfo/podinfo chart by running:

helm show values podinfo/podinfo

That's so much YAML!

Well, yes, it is. However, trust me, you will get used to it over time - many Helm charts expose a shared set of values, so navigating those values.yaml files gets easier over time. 🙂

These values can be set either by creating a file to be passed to the helm install command, or by passing them directly to the command via the --set flag. In this case, nested values can be passed using dot notation.

Naming convention

Technically, the values file can be named anything, but the convention is to name it values.yaml. This allows us to easily spot a deployment's configuration e.g. in a large Git repository.

Installing the Chart

Now, here's what we will do: We will deploy the same setup we created for podinfo manually over the course of the previous labs, but this time using a Helm chart. This will allow us to configure the whole stack all at once, using a single command.

Since we need a few values, we will create a values.yaml file to store them:

replicaCount: 3
service:
  enabled: true
  type: ClusterIP
ingress:
  enabled: true
  hosts:
    - host: helm-podinfo.127.0.0.1.nip.io
      paths:
        - path: /
          pathType: Prefix

Save the file to your workshop workspace, and we're ready to install the chart:

helm install helm-podinfo podinfo/podinfo -f values.yaml

After a short moment, the deployment will have succeeded and some instructions on how to connect to your deployed application will be printed to the console. Go and give them a try!

Lab 9: Explore the Helm installation

Have a look around - did we really manage to deploy all the manifests we created in the previous labs with a single command?

  1. Start by displaying Services, Deployments, and Ingresses
  2. Check the Ingress by connecting via the URL printed to the console
  3. Have a closer look at some of the manifests created by us vs. Helm - where do they differ?

Stretch Goal

Play around with Helm for a bit. Try to install the chart without a values.yaml file, and pass the values directly to the helm install command using the --set flag.

Try to change the number of replicas, or the ingress host. Try to install the chart again, but this time using a different release name. What happens?

Last update: September 21, 2023
Created: September 18, 2023