Introduction

You have created a Spring Boot application. It is working great on your local machine and now, you need to deploy the application somewhere else. On some platforms, you can directly submit the jar file and it will be deployed. At some places, you can spin up a virtual machine, download the source code there, build it, and run it. But, most of the time you will need to deploy the application using containers. Most of the time, Docker is used to build and run the image in a container. Also, when you upload the jar file to some platforms, the application is run inside a container under the hood.

So, in this blog, we will see 3 different ways to build a Docker image for the given Spring Boot application. Let's start:

Basic Container Image

The naive and insufficient way to build the Docker image for any application is to use a simple Dockerfile which copies the jar file inside the image and run it using java -jar command.

Create Dockerfile

Here is the Dockerfile which you can put at the root of the project:

FROM eclipse-temurin:21-jre-ubi9-minimal

ARG JAR_FILE

COPY ${JAR_FILE} application.jar

ENTRYPOINT ["java", "-jar", "/application.jar"]

We have specified one argument JAR_FILE which is the location of the jar file to use.

Building Docker Image

After creating the above Dockerfile, the below steps are used to create the Docker image:

1. Build the jar file for the Spring Boot project:
   

   ./gradlew bootJar # For Gradle build system
   

   OR
   

   ./mvnw spring-boot:build-jar # For Maven build system
   

2. Use the Dockerfile to build the Docker image using the latest jar file. In the below command replace the {IMAGE_NAME} with the required image name and {JAR_FILE} with the path to the generated jar file. The image name contains a tag as well, like - mycompany/product-service:0.0.1-SNAPSHOT:
   

   docker build --build-arg JAR_FILE={JAR_FILE} --tag {IMAGE_NAME} .
   

3. Verify if the Docker image is built using the following command. You should be able to see the image with the name specified in the command above:
   

   docker images
   

Efficient Container Image using Layered Jar

While it is possible and easy to package a Spring Boot uber jar as a Docker image (as mentioned in the previous method), there are many downsides to copying and running the fat jar as-is in the Docker image. For instance,

• There is some extra overhead when running uber jar without unpacking it.
• Putting the application's code and all of its dependencies in a single layer is not optimal.

Since we compile our code more often than upgrading the Spring Boot version, it is better to separate things a bit more. If we put those jar files (which are rarely changed) in the layer before the application layer, then Docker often needs to change only the bottom layer and can pick the rest from its cache.

Enable Layered Jar

To create a layered Docker image, we need to create a layered jar first. Nowadays, it is enabled by default in Gradle and Maven. You can enable or disable the layered jar behavior using the following setting:

// build.gradle
tasks.named("bootJar") {
    layered {
        enabled = false
    }
}

// build.gradle.kts
tasks.named("bootJar") {
   layered {
      enabled.set(false)
   }
}

org.springframework.bootspring-boot-maven-plugintrue

You can even tune how the layers are created. See the documentation for gradle or maven configuration.

Create Dockerfile

Below is the Dockerfile, which can be used to take advantage of the layered jar and to create a layered Docker image of the Spring Boot application.

# Perform the extraction in a separate builder container
FROM eclipse-temurin:21-jre-ubi9-minimal AS builder

WORKDIR /builder

# This points to the built jar file in the target folder
# Adjust this to 'build/libs/*.jar' if you're using Gradle
ARG JAR_FILE=target/*.jar

# Copy the jar file to the working directory and rename it to application.jar
COPY ${JAR_FILE} application.jar

# Extract the jar file using an efficient layout
RUN java -Djarmode=tools -jar application.jar extract --layers --destination extracted

# This is the runtime container
FROM eclipse-temurin:21-jre-ubi9-minimal

WORKDIR /application

# Copy the extracted jar contents from the builder container into the working directory in the runtime container
# Every copy step creates a new docker layer
# This allows docker to only pull the changes it really needs
COPY --from=builder /builder/extracted/dependencies/ ./
COPY --from=builder /builder/extracted/spring-boot-loader/ ./
COPY --from=builder /builder/extracted/snapshot-dependencies/ ./
COPY --from=builder /builder/extracted/application/ ./

# Start the application jar - this is not the uber jar used by the builder
# This jar only contains application code and references to the extracted jar files
# This layout is efficient to start up and CDS friendly
ENTRYPOINT ["java", "-jar", "application.jar"]

Building Docker Image

The steps to build the layered Docker image are the same as building a basic Docker image. Please refer there.

Cloud Native Buildpacks

What if I tell you that you can create a Docker image without creating a Dockerfile? We can build docker images directly from the Gralde or Maven plugin using Cloud Native Buildpacks. Some platforms (like Heroku or Cloud Foundry) use Buildpacks to convert provided jar files into runnable images.

Spring Boot includes buildpack support directly for Maven and Gradle. We don't need to include any additional plugins. Just run the below command:

./gradlew bootBuildImage # For gradle build system

OR

./mvnw spring-boot:build-image # For maven build system

The above command generates an image with the default name {PROJECT_NAME}:${PROJECT_VERSION}. If you want to configure the name of the generated image, you can follow the below steps:

Configure image name for Gradle build system

We can configure the bootBuildImage task to set the name of the image, like this:

// For build.gradle.kts
val imagePrefix = "javarush"
val dockerImageName = "docker-example"
tasks.named("bootBuildImage") {
   imageName.set("${imagePrefix}/${dockerImageName}:${version}")
}

// For build.gradle
def imagePrefix = "javarush"
def dockerImageName = "docker-example"
tasks.named("bootBuildImage") {
   imageName = "${imagePrefix}/${dockerImageName}:${version}"
}

Configure image name for Maven build system

We can configure spring-boot-maven-plugin to use another image name, like this:

javarush

...

org.springframework.bootspring-boot-maven-plugin${imagePrefix}/${project.artifactId}:${project.version}

Configure image name while running the command

We can even define the name of the image while running the command to build the image.

./gradlew bootBuildImage --imageName=javarush/docker-example:1.0.0 # For grade build system

./mvnw spring-boot:build-image -Dspring-boot.build-image.imageName=javarush/docker-example:1.0.0 # For maven build system

You can see the documentation to further configure Gradle or Maven plugin.
This is my go-to method to create a Docker image for any Spring Boot application.

Running Docker Container

Once you create a docker image, you need to make sure that it works as expected. After you make sure that the image is created, you can directly run it using the docker run command. For example,

docker run -p "8080:8080" {IMAGE_NAME}

But, this is not how images are used in production applications. Docker Compose is used to run and manage multiple docker images.

Conclusion

In this blog, we have seen how to build Docker images for Spring Boot applications using different methods. Being able to build docker images for your apps is a must skill to know because the image is what gets delivered. Thanks for reading the article till the end. I appreciate it. I will meet you in the next one. As always, all feedback and suggestions are welcome.