红帽全球峰会此内容没有您所选择的语言版本。
Chapter 11. Producing a native executable
You can produce a native executable from your Quarkus application using a container runtime such as Podman or Docker. Quarkus produces a binary executable using a builder image, which you can use together with the Red Hat Universal Base Images RHEL8-UBI and RHEL8-UBI minimal. Red Hat build of Quarkus 1.7 uses registry.access.redhat.com/quarkus/mandrel-20-rhel8:20.3 as a default for the quarkus.native.builder-image property.
The native executable for your application contains the application code, required libraries, Java APIs, and a reduced version of a virtual machine (VM). The smaller VM base improves the startup time of the application and produces a minimal disk footprint.
Procedure
Open the Getting Started project
pom.xmlfile and verify that it includes thenativeprofile:<profiles> <profile> <id>native</id> <properties> <quarkus.package.type>native</quarkus.package.type> </properties> </profile> </profiles>NoteUsing Quarkus
nativeprofile allows you to run both the native executable and the native image tests.Build a native executable using one of the following methods:
Build a native executable with Docker:
./mvnw package -Pnative -Dquarkus.native.container-build=true
Build a native executable with Podman:
./mvnw package -Pnative -Dquarkus.native.container-build=true -Dquarkus.native.container-runtime=podman
These commands create the
getting-started-*-runnerbinary in thetargetdirectory.ImportantCompiling a Quarkus application to a native executable consumes a lot of memory during analysis and optimization. You can limit the amount of memory used during native compilation by setting the
quarkus.native.native-image-xmxconfiguration property. Setting low memory limits might increase the build time.
Run the native executable:
./target/getting-started-*-runner
When you build the native executable the
prodprofile is enabled and the Quarkus native tests run using theprodprofile. You can change this using thequarkus.test.native-image-profileproperty.
11.1. Creating a container manually
This section shows you how to manually create a container image with your application for Linux X86_64. When you produce a native image using the Quarkus Native container it creates an executable that targets the Linux X86_64 operating system. If your host operating system is different from this, you will not be able to run the binary directly and you will need to create a container manually.
Your Quarkus Getting Started project includes a Dockerfile.native in the src/main/docker directory with the following content:
FROM registry.access.redhat.com/ubi8/ubi-minimal WORKDIR /work/ COPY target/*-runner /work/application RUN chmod 775 /work EXPOSE 8080 CMD ["./application", "-Dquarkus.http.host=0.0.0.0"]
The Dockerfiles use UBI as a base image. This base image was designed to work in containers. The Dockerfiles use the minimal version of the base image to reduce the size of the produced image.
Procedure
Build a native Linux executable using one of the following methods:
Build a native executable with Docker:
./mvnw package -Pnative -Dquarkus.native.container-build=true
Build a native executable with Podman:
./mvnw package -Pnative -Dquarkus.native.container-build=true -Dquarkus.native.container-runtime=podman
Build the container image using one of the following methods:
Build the container image with Docker:
docker build -f src/main/docker/Dockerfile.native -t quarkus-quickstart/getting-started .
Build the container image with Podman
podman build -f src/main/docker/Dockerfile.native -t quarkus-quickstart/getting-started .
Run the container:
Run the container with Docker:
docker run -i --rm -p 8080:8080 quarkus-quickstart/getting-started
Run the container with Podman:
podman run -i --rm -p 8080:8080 quarkus-quickstart/getting-started
For information about deploying Quarkus Maven applications on Red Hat OpenShift Container Platform, see Deploying your Quarkus applications on Red Hat OpenShift Container Platform.
