This PR enables one to create the Hadoop
release tarball on Windows, complete with
the native binaries (including winutils.exe).
This PR contains the following changes -
* Prevents splitting during array element
expansion - this is needed since we need
to pass the arguments correctly to maven.
* Install Python 3.11.8 and pip to the
Windows docker image for building
Hadoop.
* pom file changes to get maven to invoke
the releasedocmaker script through
bash.exe on Windows.
This PR gets Yetus to run on Windows 10
against the Hadoop codebase. It introduces
the following changes to allow us to setup
the nightly CI on Jenkins for Hadoop on
Windows 10.
* Hadoop personality changes for Yetus.
Additional arguments have been passed,
which are necessary to build and test
Hadoop on Windows 10.
* Docker image for building Hadoop on
Windows 10.
Installs the necessary tools that are
needed to run Yetus.
* dev-support/jenkins.sh file.
Passing of some flags are handled here
which are needed for the nightly CI.
This allows for builds to be run with options like
--mvnargs="-Dhttp.keepAlive=false -Dmaven.wagon.http.pool=false"
Contributed by Ayush Saxena.
Change-Id: I396e82d0915d679657d063a948f865041bcdde29
Since April 2022/CVE-2022-24765, git refuses to work in directories
whose owner != the current user, unless explicitly told to trust it.
This patches the create-release script to trust the /build/source
dir mounted from the hosting OS, whose userid is inevitably different
from that of the account in the container running git.
Contributed by: Steve Loughran, Ayush Saxena and the new git error messages
Removed findbugs from the hadoop build images and added spotbugs instead.
Upgraded SpotBugs to 4.2.2 and spotbugs-maven-plugin to 4.2.0.
Reviewed-by: Masatake Iwasaki <iwasakims@apache.org>
- Upgrade Yetus to 0.13.0 to support Python 3 for the release scripts.
- Removed determine-flaky-tests-hadoop.py.
- Temporarily disabled shelldocs check due to YETUS-1099.
Reviewed-by: Inigo Goiri <inigoiri@apache.org>
Reviewed-by: Mingliang Liu <liuml07@apache.org>
Use spotbugs instead of findbugs. Removed findbugs from the hadoop build images,
and added spotbugs in the images instead.
Reviewed-by: Masatake Iwasaki <iwasakims@apache.org>
Reviewed-by: Inigo Goiri <inigoiri@apache.org>
Reviewed-by: Dinesh Chitlangia <dineshc@apache.org>
This switches the SnappyCodec to use the java-snappy codec, rather than the native one.
To use the codec, snappy-java.jar (from org.xerial.snappy) needs to be on the classpath.
This comesin as an avro dependency, so it is already on the hadoop-common classpath,
as well as in hadoop-common/lib.
The version used is now managed in the hadoop-project POM; initially 1.1.7.7
Contributed by DB Tsai and Liang-Chi Hsieh
