quarkus-openshift-workshop

A workshop combining DevSpaces, Quarkus, Backstage, OpenShift, Trusted Application Pipeline, and OpenShift.ai

Running the application in dev mode (locally)

Prerequisites:

• Java 21
• Podman or Docker

Open 4 terminals and run the following commands:

• Terminal 1

cd fight-ui
./mvnw quarkus:dev

• Terminal 2 (variant OpenShift AI)

cd fight-service
export LLM_API_ENDPOINT=https://<your-openshift-ai-endpoint>/v1
export LLM_MODEL=granite-7b-instruct
./mvnw quarkus:dev

• Terminal 2 (variant Open AI)

cd fight-service
export OPENAI_API_KEY=your_api_key
./mvnw quarkus:dev 

• Terminal 3

cd hero-service
./mvnw quarkus:dev

• Terminal 4

cd villain-service
./mvnw quarkus:dev

Then, open your browser and navigate to http://localhost:8080 to see the application running (fight-UI).

Deployment on OpenShift

1. In a terminal, run the following commands:

kubectl apply -f kubernetes/frontend-route.yaml -f kubernetes/hero-database.yaml -f kubernetes/villain-database.yaml

2. You need to choose between OpenAI or OpenShift AI variant:

• To deploy the OpenAI variant, edit the kubernetes/openai.yaml and add your OpenAI API key.

  echo -n $YOUR_KEY | base64

  apiVersion: v1
  kind: Secret
  metadata:
    labels:
      app: llm
    name: llm-config
  data:
    LLM_ENDPOINT: aHR0cHM6Ly9hcGkub3BlbmFpLmNvbS92MS8=
    MODEL_NAME: Z3B0LTRvLW1pbmk=
    OPENAI_API_KEY: Copy output here
  type: Opaque

  Then apply the secret:

  kubectl apply -f kubernetes/openai.yaml
  git update-index --assume-unchanged kubernetes/openai.yaml # to make sure you do not leak the key

• To deploy the OpenShift AI variant:

  kubectl apply -f kubernetes/openshift-ai.yaml

3. Deploy the application:

kubectl apply -f fight-ui/kubernetes.yml -f hero-service/kubernetes.yml -f villain-service/kubernetes.yml -f fight-service/kubernetes.yml

4. Retrieve the URL of the frontend:

echo "Route: https://`oc get routes -o json --field-selector metadata.name=fight-ui | jq -r '.items[0].spec.host'`"