A Dojo for using Terraform and Helm together
This is a dojo to help familiarise yourself with using helm and terraform together. It is focused on running it locally on minikube. This is a dojo that aims to decrease the overwhelming details of these tools, and rather focus on the value you can get from them.
Sections
Setup a local helm chart
Setup terraform with helm locally
Run them locally on minikube
Steps
Preperation
Make sure you have thse installed locally:
- helm cli
- terraform cli
- minikube
Make sure you have minikube running
$ minikube status
Make sure you are able to run kubectl in the context of your minikube
Setup a local helm chart
First, create a folder called "charts".
$ mkdir charts
You can create a helm chart from a template by running:
$ helm create ./charts/example
Take a look at the chart content
Some useful documentation:
https://helm.sh/docs/topics/charts/
https://helm.sh/docs/chart_template_guide/getting_started/
A high level overview of what you should see
./charts/example/
Chart.yaml # A YAML file containing information about the chart
values.yaml # The default configuration values for this chart
charts/ # A directory containing any charts upon which this chart depends.
templates/ # A directory of templates that, when combined with values,
# will generate valid Kubernetes manifest files.
templates/NOTES.txt # OPTIONAL: A plain text file containing short usage notes
Inside templates remove all the files and folders except for the _helpers.tpl file
./charts/example/templates should now be an empty folder with just one file, _helpers.tpl
The _helpers.tpl file has your reusable values
_helpers.tpl: A place to put template helpers that you can re-use throughout the chart
Copy your kubernetes *.yaml files over to the templates folder (not the values.yaml file - but all the others)
./charts/example/templates
_helpers.tpl
deployment.yaml
namespace.yaml
service.yaml
Replace the values.yaml with the value.yaml in the chart root folder, ./charts/example/values.yaml
cat ./charts/example/values.yaml <-- shoulld now have your values inside
Test your chart with these commands:
First check for any syntax issues:
$ helm lint ./charts/example
You can test in --dry-run mode:
$ helm install example-chart --dry-run --debug ./charts/example
Then you can actially install your chart on your kubernetes environment by running:
$ helm install example-chart ./charts/example
To remove your chart again you can run:
$ helm delete example-chart
To list your charts run:
$ helm list -aq
Check your new resources on minikube:
$ kubectl get all -n example
Test your service and application:
$ kubectl port-forward service/example 8001:80 -n example
Open up on your browser: http://localhost:8001/
Before continuing please delete the chart:
$ helm delete example-chart
Setup with terraform
Why use terraform with helm? Terraform helps manage resources for day 2 problems. When you have tons of applications running on a cluster across many namespaces, it's easy to get architectural drift. What your configuration files define and what is actually on the system start to drift, and there ends up being resources that run on kubernetes that you "loose". Terraform helps with this part.
Some useful docs
https://registry.terraform.io/providers/hashicorp/helm/latest/docs
https://learn.hashicorp.com/tutorials/terraform/helm-provider
Copy the main.tf file from the ./use-files/terraform folder to the root of this dojo exercise
./main.tf
Initialise your terraform:
$ terraform init
Test terraform
$ terraform plan
$ terraform apply
yes
Check your kubernetes resources on your minikube environment:
Check your new resources on minikube:
$ kubectl get all -n example
Test your service and application:
$ kubectl port-forward service/example-chart 8001:80 -n example
Open up on your browser: http://localhost:8001/
Then when you are done you can destroy all the resources you created:
$ terraform destroy
yes