Workshop Runtimes Helm Chart

A Helm chart for provisioning OpenShift resources needed for workshop environments. This chart deploys all required runtime resources into <user>-devspaces namespaces, bypassing typical access restrictions.

Overview

While primarily designed for DevSpaces workshop environments, this chart can provision any cluster resources needed for workshop participants. Resources are deployed into user-specific namespaces following the pattern <user>-devspaces.

What's Included

This chart provisions the following resources:

• MinIO: S3-compatible object storage (API on port 9000, console on 9001)
• Kafka: Strimzi-managed Kafka cluster with auto-topic creation
• Kafka Topics: Pre-configured topics for workshop exercises
• Kafka UI: Web-based Kafka management interface
• KEDA: Kubernetes Event-Driven Autoscaling for Kafka topics
• Tekton Pipelines: CI/CD pipelines for Spring Boot applications
  • Maven build and package
  • Container image build
  • Application deployment
• Tekton Triggers: Automated pipeline triggers for git pushes
• Routes: OpenShift routes for API and console access
• RBAC: Necessary service accounts and role bindings

Usage

Installation

Option 1: Using the Published Helm Repository (Recommended)

First, add the workshop Helm repository to your OpenShift cluster:

oc apply -f https://raw.githubusercontent.com/na-launch-workshop/platform-workshop-runtimes/master/install.yaml

Then install the chart to a user's namespace:

helm install <release-name> workshop-charts/workshop-runtimes -n <user>-devspaces

For example:

helm install runtimes workshop-charts/workshop-runtimes -n alice-devspaces

Option 2: Install Directly from Source

Clone the repository and deploy the chart:

git clone https://github.com/na-launch-workshop/platform-workshop-runtimes.git
cd platform-workshop-runtimes
helm install <release-name> . -n <user>-devspaces

Upgrade

Update an existing deployment:

# From the published repository
helm upgrade <release-name> workshop-charts/workshop-runtimes -n <user>-devspaces

# Or from local source
helm upgrade <release-name> . -n <user>-devspaces

Uninstall

Remove all provisioned resources :

helm uninstall <release-name> -n <user>-devspaces

Configuration

Edit values.yaml to customize the deployment:

replicaCount: 1

image:
  repository: quay.io/minio/minio
  tag: latest
  pullPolicy: IfNotPresent

env:
  rootUser: admin
  rootPassword: admin123

service:
  type: ClusterIP
  ports:
    api: 9000
    console: 9001

Key Configuration Options

• MinIO Credentials: Change env.rootUser and env.rootPassword
• Image Version: Update image.tag for specific MinIO versions
• Resource Limits: Add resource requests/limits in resources
• Node Placement: Configure nodeSelector, tolerations, affinity

Adding New Resources

To add new workshop resources:

1. Create a new template file in templates/ (e.g., templates/myresource.yaml)
2. Use Helm templating for namespace injection: namespace: {{ .Release.Namespace }}
3. Add configuration values to values.yaml if needed
4. Deploy with helm upgrade

Tekton Pipelines

The chart includes a complete CI/CD stack for Spring Boot applications:

Pipeline: springboot-build-deploy

1. maven-package: Clones git repo and runs Maven build
2. build-image: Creates container image using Buildah
3. deploy-app: Deploys to OpenShift with the image

Triggers

Automated triggers respond to git webhooks and start pipelines automatically:

• workshop-springboot-hello_by_lang - Standard Spring Boot app
• workshop-springboot-using_camel - Camel integration app

Architecture

<user>-devspaces namespace
├── MinIO (Object Storage)
│   ├── API Service (:9000)
│   └── Console Service (:9001)
├── Kafka Cluster
│   ├── Kafka Broker (:9092)
│   ├── Zookeeper
│   └── Topics (auto-created)
├── Kafka UI (Management)
├── KEDA ScaledObjects
└── Tekton CI/CD
    ├── Pipeline Definitions
    ├── Task Definitions
    ├── Triggers (EventListener)
    └── PersistentVolumeClaims

Publishing

This chart is automatically published to GitHub Pages on every push to the master branch via GitHub Actions. The workflow:

1. Packages the Helm chart
2. Generates the Helm repository index
3. Publishes to https://na-launch-workshop.github.io/platform-workshop-runtimes

Users can then install via the HelmChartRepository CR in install.yaml.

Development

Chart Structure

.
├── Chart.yaml              # Chart metadata
├── values.yaml             # Default configuration values
├── install.yaml            # HelmChartRepository CR for OpenShift
├── templates/              # Kubernetes resource templates
│   ├── deployment.yaml     # MinIO deployment
│   ├── service.yaml        # MinIO services
│   ├── route-*.yaml        # OpenShift routes
│   ├── kafka.yaml          # Kafka cluster
│   ├── kafka-topics.yaml   # Topic definitions
│   ├── kafka-ui.yaml       # Kafka UI
│   ├── keda-*.yaml         # KEDA autoscaling
│   └── tekton-*.yaml       # CI/CD pipelines
└── .github/
    └── workflows/
        └── helm-release.yaml  # Auto-publish workflow

Testing Changes

1. Make changes to templates or values
2. Validate with helm template . --debug
3. Test deployment in a dev namespace
4. Verify resources are created: oc get all -n <namespace>
5. Push to master to trigger automatic publishing to the Helm repository

Common Tasks

Check Deployed Resources

helm list -n <user>-devspaces
oc get all -n <user>-devspaces

View MinIO Console

oc get route console -n <user>-devspaces

Monitor Tekton Pipelines

tkn pipeline list -n <user>-devspaces
tkn pipelinerun logs -f -n <user>-devspaces

Access Kafka

oc get kafka -n <user>-devspaces
oc get kafkatopic -n <user>-devspaces

Troubleshooting

Resources Not Creating

Check Helm deployment status:

helm status <release-name> -n <user>-devspaces

Pipeline Failures

View pipeline run logs:

tkn pipelinerun ls -n <user>-devspaces
tkn pipelinerun logs <pipelinerun-name> -n <user>-devspaces

Kafka Issues

Check Kafka operator logs:

oc logs -l name=strimzi-cluster-operator -n openshift-operators

Notes

• Resources use ephemeral storage by default (suitable for workshops)
• Kafka topics are auto-created when accessed
• The chart assumes Strimzi Kafka operator is installed cluster-wide
• Tekton operator must be installed for pipeline resources
• KEDA operator required for autoscaling features

License

This chart is for workshop/educational use.