Skip to main content

Building optimized OCI images with Jib

Objective​

In this lab, you will learn a straightforward and efficient method for optimizing images from the Open Container Initiative (OCI) using Jib.

Prerequisites​

Context​

The OCI currently contains three specifications: the Runtime Specification, the Image Specification and the Distribution Specification. Jib is a tool to build optimized OCI images without using a container runtime. It's available as a Maven or Gradle-plugin as well as a Java library. In our case we're going to use the Maven-plugin in order to create an optimized container image.

The important part for the build with Jib can be found in the pom.xml-file in the plugins section:

<plugin>
    <groupId>com.google.cloud.tools</groupId>
    <artifactId>jib-maven-plugin</artifactId>
    <version>3.4.0</version>
    <configuration>
        <from><image>public.ecr.aws/docker/library/amazoncorretto:17.0.9-alpine3.18</image></from>
        <container>
            <user>1000</user>
        </container>
    </configuration>
</plugin>

At this stage, the configuration is quite compact. We simply specify that the Java process in the container image will not run as the root user, but instead with a different user ID. Additionally, we opt for the Alpine Linux base image to enhance the overall performance.

1. Changing the source code and pushing the image​

You are now going to change the application code of unicorn-store-spring/src/main/java/com/unicorn/store/controller/UnicornController.java to identify the new version of the application deployment.

Change the contents of the getWelcomeMessage function to identify the new version of the application:

@GetMapping("/")
public ResponseEntity<String> getWelcomeMessage() {
    return new ResponseEntity<>("Welcome to the Unicorn Store - from Jib generated Image!", HttpStatus.OK);
}
info

AWS Cloud9 does not auto-save your files. Please ensure to save your files before deploying any changes via Ctrl+S or the top menu File→Save all.

info

AWS Cloud9 does not auto-save your files. Please ensure to save your files before deploying any changes via Ctrl+S or the top menu File→Save all.

Build new container images using mvn compile jib:build:

export ECR_URI=$(aws ecr describe-repositories --repository-names unicorn-store-spring | jq --raw-output '.repositories[0].repositoryUri')
aws ecr get-login-password --region $AWS_REGION | docker login --username AWS --password-stdin $ECR_URI
cd ~/environment/unicorn-store-spring
IMAGE_TAG=i$(date +%Y%m%d%H%M%S)
IMAGE_PATH=$ECR_URI:$IMAGE_TAG
mvn compile jib:build -Dimage=$IMAGE_PATH

IMAGE_PATH=$ECR_URI:latest
mvn compile jib:build -Dimage=$IMAGE_PATH

2. Re-deploying the application​

After pushing the new image to ECR, you can re-trigger the deployment of the application:

kubectl rollout restart deploy unicorn-store-spring -n unicorn-store-spring
kubectl rollout status deployment unicorn-store-spring -n unicorn-store-spring

3. Testing the application​

Run the following API call to verify that the new version of the application was successfully deployed:

export SVC_URL=http://$(kubectl get svc unicorn-store-spring -n unicorn-store-spring -o json | jq --raw-output '.status.loadBalancer.ingress[0].hostname')
curl --location --request GET $SVC_URL'/' --header 'Content-Type: application/json'; echo

jib-result

4. Retrieving the results​

  1. Verify the new application image size in the Amazon ECR console:

jib-ecr

  1. Retrieve the logs for the application as outlined in the previous section. Below you can find an example starting time after the optimization of the Amazon EKS deployment:
kubectl logs $(kubectl get pods -n unicorn-store-spring -o json | jq --raw-output '.items[0].metadata.name') -n unicorn-store-spring
kubectl logs $(kubectl get pods -n unicorn-store-spring -o json | jq --raw-output '.items[0].metadata.name') -n unicorn-store-spring | grep "Started StoreApplication"

jib-eks

Conclusion​

As you can see, we managed to decrease the container image size from 380 MB to 212 MB, resulting in a reduction of approximately 45% without making any changes to the code. This was achieved by using Jib and the Linux Alpine base image for container creation. Additionally, the application startup time improved by around 2 seconds.