title | slug | license |
---|---|---|
How to build Apache Gravitino |
/how-to-build |
This software is licensed under the Apache License version 2. |
- Linux or macOS operating system
- Git
- A Java Development Kit, version 8 to 17, installed in your environment to launch Gradle
- Python 3.8, 3.9, 3.10, or 3.11 to build the Gravitino Python client
- Optionally, Docker to run integration tests
:::info Please read the following notes before trying to build Gravitino.
- Gravitino requires a minimum of JDK8 and supports up to JDK17 to run Gradle, so you need to install a JDK, version 8 to 17, to launch the build environment.
- Gravitino itself supports using JDK 8, 11, or 17 to build. The Gravitino Trino connector uses JDK17 to build (to avoid vendor-related issues on some platforms, Gravitino uses the specified Amazon Corretto OpenJDK 17 to build the Trino connector on macOS). You don't have to preinstall the specified JDK environment, as Gradle detects the JDK version needed and downloads it automatically.
- Gravitino uses the Gradle Java Toolchain to detect and manage JDK versions, and it checks the installed JDK by running the
./gradlew javaToolchains
command. See Gradle Java Toolchain. - Gravitino excludes all Docker-related tests by default. To run Docker-related tests, make sure you have installed Docker in your environment and either (1) set
skipDockerTests=false
in thegradle.properties
file (or use-PskipDockerTests=false
in the command) or (2)export SKIP_DOCKER_TESTS=false
in the shell. Otherwise, all tests requiring Docker will be skipped. - macOS uses
docker-connector
to make the Gravitino Trino connector work with Docker for macOS. See docker-connector,$GRAVITINO_HOME/dev/docker/tools/mac-docker-connector.sh
, and$GRAVITINO_HOME/dev/docker/tools/README.md
for more details. - You can use OrbStack as a replacement for Docker for macOS. See OrbStack. With OrbStack, you can run Gravitino integration tests without needing to install
docker-connector
. :::
-
Clone the Gravitino project.
If you want to contribute to this open-source project, please fork the project on GitHub first. After forking, clone the forked project to your local environment, make your changes, and submit a pull request (PR).
git clone [email protected]:apache/gravitino.git
-
Build the Gravitino project. Running this for the first time can take 15 minutes or more.
cd gravitino ./gradlew build
The default specified JDK version is 8, but if you want to use JDK 11 or 17 to build, modify the property
jdkVersion
to 11 or 17 in thegradle.properties
file, or specify the version with-P
, like:./gradlew build -PjdkVersion=11
Or:
./gradlew build -PjdkVersion=17
The ./gradlew build
command builds all the Gravitino components, including the Gravitino server, Java and Python clients, Trino and Spark connectors, and more.
For the Python client, the ./gradlew build
command builds the Python client with Python 3.8 by default. If you want to use Python 3.9, 3.10, or 3.11 to build, please modify the property pythonVersion
to 3.9, 3.10, or 3.11 in the gradle.properties
file, or specify the version with -P
like:
./gradlew build -PpythonVersion=3.9
Or:
./gradlew build -PpythonVersion=3.10
Or:
./gradlew build -PpythonVersion=3.11
If you want to build a module on its own, like the Spark connector, you can use Gradle to build a module with a specific name, like so:
./gradlew spark-connector:spark-runtime-3.4:build -PscalaVersion=2.12
This creates gravitino-spark-connector-runtime-{sparkVersion}_{scalaVersion}-{version}.jar
under the spark-connector/v3.4/spark-runtime/build/libs
directory. You could replace 3.4
with 3.3
or 3.5
to specify different Spark versions and replace 2.12
with 2.13
for different Scala versions. The default Scala version is 2.12
if -PscalaVersion
is not specified.
:::info Gravitino Spark connector doesn't support Scala 2.13 for Spark 3.3. :::
:::note The first time you build the project, downloading the dependencies may take a while.
You can add -x test
to skip the tests using ./gradlew build -x test
.
The built Gravitino libraries are Java 8 compatible and verified under the Java 8, 11, and 17 environments. You can use Java 8, 11, or 17 runtimes to run the Gravitino server, no matter which JDK version was used to build the project.
The built jars are under the modules build/libs
directory. You can publish them in your Maven repository for use in your project.
:::
-
Get the Gravitino server binary package.
./gradlew compileDistribution
The compileDistribution
command creates a distribution
directory in the Gravitino root directory.
:::note
The ./gradlew clean
command deletes the distribution
directory.
:::
-
Assemble the Gravitino server distribution package.
./gradlew assembleDistribution
The assembleDistribution
command creates gravitino-{version}-bin.tar.gz
and gravitino-{version}-bin.tar.gz.sha256
under the distribution
directory.
You can deploy these to your production environment.
:::note
The gravitino-{version}-bin.tar.gz
file is the Gravitino server distribution package, and the gravitino-{version}-bin.tar.gz.sha256
file is the sha256 checksum file for the Gravitino server distribution package.
:::
- Assemble the Gravitino Trino connector package
./gradlew assembleTrinoConnector
or
./gradlew assembleDistribution
This creates gravitino-trino-connector-{version}.tar.gz
and
gravitino-trino-connector-{version}.tar.gz.sha256
under the distribution
directory. You can uncompress and deploy it to Trino to use the Gravitino Trino connector.
- Assemble the Gravitino Iceberg REST server package
./gradlew assembleIcebergRESTServer
This creates gravitino-iceberg-rest-server-{version}.tar.gz
and gravitino-iceberg-rest-server-{version}.tar.gz.sha256
under the distribution
directory. You can uncompress and deploy it to use the Gravitino Iceberg REST server.
On Windows:
Refer to this guide for installation: WSL Installation Guide
Note: Gravitino can run successfully on Ubuntu 22.04
This step involves setting up your Windows machine's Windows Subsystem for Linux (WSL). WSL allows you to run a Linux distribution alongside Windows, providing a Linux-like environment for development.
On Ubuntu (WSL):
sudo apt update
sudo apt install apt-transport-https ca-certificates curl software-properties-common
Updating the package list ensures you have the latest information on the newest versions of packages and dependencies. Installing the necessary packages lets your system download and manage additional software securely.
On Ubuntu (WSL):
-
Edit your
~/.bashrc
file using any editor. Here,vim
is used:vim ~/.bashrc
-
Add the following lines at the end of the file. Replace
/usr/lib/jvm/java-11-openjdk-amd64
with your actual Java installation path:export JAVA_HOME="/usr/lib/jvm/java-11-openjdk-amd64" export PATH=$PATH:$JAVA_HOME/bin
-
Save and quit in vim using
:wq
. -
Run
source ~/.bashrc
to update your shell session's environment variables.Editing the
~/.bashrc
file allows you to set environment variables available in every terminal session. SettingJAVA_HOME
and updatingPATH
ensures that your system uses the correct Java version for development.
On Ubuntu (WSL):
curl -fsSL https
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable"
sudo apt update
sudo apt install docker-ce
sudo service docker start
sudo docker run hello-world
sudo usermod -aG docker $USER
These commands install Docker. Running hello-world
verifies the installation. Adding your user to the Docker group allows you to run Docker commands without sudo
.
On Ubuntu (WSL):
sudo apt update
sudo apt install software-properties-common
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt update
sudo apt install python3.11
python3.11 --version
These commands add a repository that provides the latest Python versions and installs Python 3.11.
On Ubuntu (WSL):
git clone https://github.com/apache/gravitino.git
cd gravitino
./gradlew compileDistribution -x test
cd distribution/package/
./bin/gravitino.sh start
Access http://localhost:8090
Building the Gravitino project compiles the necessary components, and starting the server allows you to access the application in your browser.
Please refer to CONTRIBUTING.md for instructions on running the project using VSCode or IntelliJ on Windows.