dbuchko / spring-sensors-rabbit

Demo application for VMware Tanzu Application Platform

An application to demonstrate the capabilities of VMware Tanzu Application Platform.

Prerequisites

To deploy this demo app you need to have VMware Tanzu Application Platform with dev, light or full profile installed. For installation instructions, please see the platform documentation. You also need to have prepared the "defult" namespace as described in the "Set Up Developer Namespaces to Use Installed Packages" section of the documentation.

This application acts as a consumer of sensor data which the application stores in an in-memory database and displays on a dashboard.

Spring Cloud Stream, a framework built on top of Spring Boot and Spring Integration, is used as a flexible messaging abstraction. Spring Cloud Stream supports a variety of binder implementations. In this case, we are using the one for RabbitMQ.

So, as a prerequisite to run this application with VMware Tanzu Application Platform, you need a RabbitMQ cluster running in the same Kubernetes namespace (e.g. provisioned via the RabbitMQ Cluster Operator for Kubernetes).

apiVersion: rabbitmq.com/v1beta1
kind: RabbitmqCluster
metadata:
  name: rmq-1

To apply this RabbitmqCluster run the following command:

kubectl apply -f rabbit/cluster.yaml

Binding an application workload to a backing service such as a RabbitMQ queue is one of the most important use cases within the context of the VMware Tanzu Application Platform. This use case is made possible by the Service Binding Specification for Kubernetes. With the service binding that is defined in the workload.yaml, the credentials that are required for the connection to the RabbitMQ cluster are magically injected as environment variables into the container.

You do need a ClusterRole that allows the "Services Toolkit" component access to bind to the RabbitMQ cluster you just created. Apply the ClusterRole definition using:

kubectl apply -f rabbit/rbac.yaml

Deploying the application

NOTE: The provided config/workload.yaml file uses the Git URL for a repo generated with this accelerator. When you want to modify the source, you must push the code to your own Git repository and then update the spec.source.git information in the config/workload.yaml file.

To deploy this application on VMware Tanzu Application Platform, execute the following command:

tanzu apps workload create spring-sensors -f config/workload.yaml

You can access the application's UI using the URL shown by running the following command (provided you have DNS configured for Cloud Native Runtimes):

tanzu app workload get spring-sensors

Deploying the publisher

The sensor data is generated and sent by this application via asynchronous messaging.

To build this application using the build service apply the an Image definition using:

kubectl apply -f demo/publisher-image.yaml

Check the build pod progress using:

kubectl get pod -l=image.kpack.io/image=spring-sensors-publisher

Once the init containers finish and the pod status shows Complete, you can check the image status using:

kubectl get image.kpack.io spring-sensors-publisher

That should show READY as True and the LATESTIMAGE populated.

We are now ready to deploy the publisher app using a Kubernetes Deployment, run this command:

kubectl apply -f demo/publisher-deployment.yaml

You can scale up the number of sensors displayed in the demo app by using:

kubectl scale deployment/spring-sensors-publisher --replicas=5