Deploy a Spring application connected to YugabyteDB Managed on Google Kubernetes Engine (GKE) by following the steps below.

This example uses the PetClinic application, connected to YugabyteDB Managed and containerized using Docker; refer to Connect a Spring Boot application.

Prerequisites

Before starting, you need to verify that the following are installed and configured:

• Google Cloud Platform (GCP) account

• GCP Cloud SDK

  • For more information, refer to Cloud SDK.

• Docker

• Your containerized Spring Boot application

  • For more information, refer to Connect a Spring Boot application

Deploy the application image to GKE

Create a repository in GCP

1. Go to https://console.cloud.google.com/ and sign in to GCP.

2. Type “artifact registry” in the search bar and select Artifact Registry.

3. Click Enable for the Artifact Registry API.

4. Click Create Repository.

5. Give your repository a name, make sure Docker is the selected format, and choose a region. Then click Create.

6. Got to your new repository and copy its URL.

Tag your application image in Docker

On your computer, do the following:

1. Configure Docker to use gcloud for authentication.

   $ gcloud auth configure-docker [region]-docker.pkg.dev
   

   Replace [region] with the region you selected in GCP.

   Adding credentials for: us-west1-docker.pkg.dev
   After update, the following will be written to your Docker config file
   located at [/Users/gavinjohnson/.docker/config.json]:
   {
     "credHelpers": {
       "us-west1-docker.pkg.dev": "gcloud"
     }
   }
   
   Do you want to continue (Y/n)?  y
   
   Docker configuration file updated.
   

2. Tag your PetClinic image with your Artifact Registry repo.

   $ docker tag spring-petclinic:latest [repo_url]/spring-petclinic:latest
   

   Replace [repo_url] with the URL of your repository.

3. Push your PetClinic image to your repo in GCP.

   $ docker push [repo_url]/spring-petclinic:latest
   

   Replace [repo_url] with the URL of your repository.

   1dc94a70dbaa: Pushed
   0d29ec96785e: Pushed
   888ed16fa8d4: Pushed
   ...
   

Go to your repo in the GCP Artifact Registry to view the image you just pushed.

Deploy the image to GKE

In GCP, do the following:

1. Type “gke” in the search bar and select Kubernetes Engine.

2. Click Enable for the Kubernetes Engine API.

3. Click Create.

4. Click Configure under Autopilot.

5. Enter a Name, select a Region, and click Create.

6. After your cluster has been created, select Workloads in the left navigation panel.

7. Click Deploy.

8. Select Existing Container Image and click Select in the Image Path.

9. Click Artifact Registry, expand the folders, select the image under spring-petclinic, and click Select.

10. Click Add Environment Variable.

11. Enter the following:

    Key: JAVA_OPTS
    Value: -Dspring.profiles.active=yugabytedb \
    -Dspring.datasource.url=jdbc:postgresql://[host]:[port]/petclinic?load-balance=true \
    -Dspring.datasource.initialization-mode=never
    

    Replace [host] and [port] with the host and port number of your YugabyteDB Managed cluster. To obtain your cluster connection parameters, sign in to YugabyteDB Managed, select your cluster and navigate to Settings.

12. Click Continue.

13. Enter a Name for your application and click Deploy (all other options should be auto-filled).

14. After your workload is deployed, click Expose on the Deployment Details.

15. Set Port to 80, Target Port to 8080, and Service Type to Load Balancer, then click Expose.

    After the load balancer is created, the Service Details page displays.

16. Click the External Endpoints link (http://34.105.123.45/) to navigate to the PetClinic application.

The PetClinic sample application is now connected to your YugabyteDB Managed cluster and running on Kubernetes on GKE.