This image builds a runtime ZIP file containing a cut-down JVM based on Amazon Corretto for arm64. It is designed to work with a ShadowJar containing an http4k AWS Lambda function.
Basic usage for an example project. Arguments to the docker run command:
- UBERJAR_PATH = path the UberJar. This is loaded from inside the container at source/<UBERJAR_PATH>
- OUTPUT_ZIP_PATH (optional) = path to the output Zipfile. Defaults to source/function.zip
./gradlew shadowJar
docker run -v $(pwd):/source http4k/amazoncorretto-lambda-runtime-arm64:latest <UBERJAR_PATH> <OUTPUT_ZIP_PATH?>The Docker image itself includes 2 script files loaded from this repo:
- bootstrap - the standard bootstrap file used by the Aws Lambda runner. Detects and passes the lambda memory settings to the JVM.
- packageZip - takes the input ShadowJar from the location in the source directory, uses
jlinkto create a streamlined version of the Java runtime based on the required modules detected, and creates the output ZIP package.
Images are hosted in DockerHub