Sensor Data Service

This service simulates real-world vehicle sensors for demonstration purposes only. In a real-world scenario, you would connect directly to actual vehicle sensors rather than using this simulation service.

Why We're Using Simulation:

• Created for demonstration purposes due to lack of access to real vehicle sensors
• In production, this service would be replaced with direct integration to actual vehicle telemetry systems

Important Security Notice:

• This demo uses embedded private keys for simplicity
• These keys could be exposed publicly, creating security risks

Set up Kubernetes Cluster

1. Add Camunda Helm Repository

helm repo add camunda https://helm.camunda.io
helm repo update

2. Start Minikube

Start a Minikube cluster with sufficient resources. You may adjust the values based on your machine/driver:

minikube start --memory=8192 --cpus=4 --driver=docker

3. Install Camunda via Helm

Install Camunda into the camunda namespace (or another of your choice) using your custom configuration file:

helm install camunda camunda/camunda-platform \
  -n proving-system --create-namespace \
  -f camunda-platform/camunda-platform-core-kind-values.yaml

Notes:

• If you do not use the default namespace, you must always include -n camunda (or your chosen namespace) in kubectl commands.
• The YAML configuration file should be available locally or from your cloud storage. Adjust the path if needed.

Sidenote: to uninstall use

helm uninstall camunda -n proving-system

4. Verify Camunda Installation

Wait a few minutes for all services to initialize. Then check the status:

kubectl get pods -n proving-system

You should see multiple Camunda components like camunda-zeebe, camunda-operate, and others showing STATUS: Running.

If some show Pending or ContainerCreating, wait until they are fully up.

5. Port Forward Zeebe Gateway for Local Access

To connect the Camunda Modeler to Zeebe, forward the gateway port:

kubectl port-forward svc/camunda-zeebe-gateway 26500:26500 -n proving-system

Keep this terminal open while deploying models from the Camunda Modeler.

6. Deploy BPMN Models from Camunda Modeler

In the Camunda Modeler:

1. Open your BPMN file.

2. Select Camunda 8 → Self-Managed.

3. Use the following connection settings:

• Zeebe Gateway Address: localhost:26500

• Authentication: None

4. Click Deploy Current Diagram.

5. Start process instances (only after you deployed the other Microservices to the Cluster).

7. Access Camunda Operate

To view and manage process instances, forward the Camunda Operate service:

kubectl port-forward svc/camunda-operate 8081:80 -n proving-system

Then open your browser at: http://localhost:8081

8. Build and Load Images in Minikube

Before deploying, make sure to build your images inside Minikube's Docker context (we don't need to upload to DockerHub with Minikube, we keep the images local):

Kubernetes does not run source code directly, it runs containers that are created from Docker images. That means every time you create a new service or want to deploy one, you first need to build its Docker image.

Make sure that the commands have to be executed in the right folder of the dedicated repository.

If you use a Macbook that is based on an ARM64 architecture ((M1, M2, M3, M4) make sure to add --platform=linux/amd64 to build your proving-service Dockerfile (see below). The Risc Zero Toolchain doesn't support linux/aarch64 but only can installed on linux/amd64 platforms.

eval $(minikube docker-env)
docker build -t sensor-data-service:latest .
docker build -t camunda-service:latest .
docker build --platform=linux/amd64 -t proving-service:latest .

Then apply your Kubernetes YAML files (you might need to switch folders):

kubectl apply -f sensor-deployment.yaml -n proving-system
kubectl apply -f camunda-service-deployment.yaml -n proving-system
kubectl apply -f proving-service.yaml -n proving-system

Always use -n proving-system (or the name that you chose) if you are working outside the default namespace.

9. Monitor Logs and Status

kubectl get pods -n proving-system
kubectl logs deployment/camunda-service -n proving-system
kubectl logs deployment/sensor-data-service -n proving-system
kubectl logs deployment/proving-service -n proving-system

10. Update Docker Images After Code Changes

If you change the code in one of your services (e.g., the Camunda worker or sensor service), you must rebuild the corresponding Docker image.

Kubernetes uses the image specified in your deployment YAML, so the container will not reflect your code changes unless the image has been updated.

To rebuild the image and apply changes in Minikube:

1. Re-enter the Minikube Docker context:

   eval $(minikube docker-env)

2. Rebuild your image(s), for example:

   docker build -t camunda-service:latest ./camunda-service

3. Restart the deployment to use the updated image:

   kubectl rollout restart deployment camunda-service -n proving-system

11. Delete deployments or services

kubectl delete deployment proving-service -n proving-system
kubectl delete service proving-service -n proving-system