...
- Check if the release year for Web UI footer is updated (the property
<release-year>
inhadoop-project/pom.xml
). If not, create a JIRA to update the property value to the right year, and propagate the fix from trunk to all necessary branches. Consider the voting time needed before publishing, it's better to use the year of (current time + voting time) here, to be consistent with the publishing time. - In JIRA, ensure that only issues in the "Fixed" state have a "Fix Version" set to release X.Y.Z.
- Verify that $HOME/.gpg defaults to the key listed in the KEYS file.
For the Apache release, a machine capable of running Docker- and Internet- capable, build the release candidate with
create-release
. Unless the--logdir
is given, logs will be in thepatchprocess/
directory. Artifacts will be in the target/artifacts NOTE: This will take quite a while, since it will download and build the entire source tree, including documentation and native components, from scratch to avoid maven repository caching issues hiding issues with the source release. dev-support/bin/create-release --asfrelease --docker --dockercacheNo Format For x86 architecture CPU, we can use the following command:
No Format dev-support/bin/create-release --asfrelease --docker --dockercache
Note:
if we encounter a 401 permission issue, it may be due to the local Maven configuration not being read. We can try the following command.
Code Block dev-support/bin/create-release --asfrelease --docker --dockercache --mvnargs="-Duser.home=<mvn setting.xml path>"
For Arm architecture CPU, we can use the following command:
Code Block dev-support/bin/create-release --docker --dockercache --mvnargs="-Dhttp.keepAlive=false -Dmaven.wagon.http.pool=false" --deploy --native --sign
Note:
For compiling on ARM architecture CPU, we can refer to the https://github.com/apache/hadoop-release-support project.While it should fail
create-release
if there are issues, doublecheck the rat log to find and fix any potential licensing issues.No Format grep 'Rat check' patchprocess/mvn_apache_rat.log
- Check that release files look ok - e.g. install it somewhere fresh and run examples from tutorial, do a fresh build, read the release notes looking for WARNINGs, etc.
- Set environment variable version for later steps.
export version=X.Y.Z-RCN
Tag the release candidate:
No Format git tag -s release-$version -m "Release candidate - $version"
- Push branch-X.Y.Z and the newly created tag to the remote repo.
(This is for branch-2.10 only.
mvn deploy
is invoked by the dev-support/bin/create-release script of 3.0.0 and above. see HADOOP-15058.)
Deploy the maven artifacts, on your personal computer. Please be sure you have completed the prerequisite step of preparing thesettings.xml
file before the deployment. You might want to do this in private and clear your history file as your gpg-passphrase is in clear text.No Format mvn deploy -Psign,dist -DskipTests
Copy release files to a public place and ensure they are readable. Note that
home.apache.org
only supports SFTP, so this may be easier with a graphical SFTP client like Nautilus, Konqueror, etc.No Format sftp home.apache.org > cd public_html > mkdir hadoop-${version} > put -r /home/hadoop/hadoop-${version} .... > bye
- Log into Nexus, select "
Staging
Repositories" from the left navigation pane, select the check-box against the specific hadoop repository, andclose
the release. At the time of deploying the maven artifacts, if there are different hadoop items of this release candidate at “Staging Repositories”, drop the stale RC first. - Call a release vote on common-dev at hadoop.apache.org. It's usually a good idea to start the release vote on Monday so that people will have a chance to verify the release candidate during the week. Example
- If the release candidate contains a serious issue, withdraw the vote, make necessary changes, and repeat this process.
- If non-trivial changes are committed to the release branch, ensure the commits are present in the upstream branches.
...
- In JIRA, "release" the version, setting the date to the end-of-vote date. Visit the "Administer Project" page, then the "Manage versions" page. You need to have the "Admin" role in HADOOP, HDFS, MAPREDUCE, and YARN.
- Set environment variable version for later steps.
export version=X.Y.Z
Tag the release. Do it from the release branch and push the created tag to the remote repository:
No Format git tag -s rel/release-${version} -m "Hadoop ${version} release" git push origin rel/release-${version}
- Copy release files to the distribution directory
Check out the corresponding svn repo if need be
No Format svn co https://dist.apache.org/repos/dist/release/hadoop/common/ hadoop-dist
- Copy the release files to hadoop-dist/hadoop-${version}
- Update the symlinks to current2 and stable2. The release directory usually contains just two releases, the most recent from two branches.
Commit the changes (it requires a PMC privilege)
No Format svn ci -m "Publishing the bits for release ${version}"
- Usually binary tarball becomes larger than 300MB, so it cannot be directly uploaded to the distribution directory. Use the dev directory (https://dist.apache.org/repos/dist/dev/hadoop/) first and then move it to the distribution directory by
svn move
.
- Update upstream branches to make them aware of this new release:
Copy and commit the CHANGELOG.md and RELEASENOTES.md:
No Format cp target/artifacts/RELEASENOTES.md hadoop-common-project/hadoop-common/src/site/markdown/release/${version}/RELEASENOTES.${version}.md cp target/artifacts/CHANGELOG.md hadoop-common-project/hadoop-common/src/site/markdown/release/${version}/CHANGELOG.${version}.md
Copy the jdiff xml files for this version to their appropriate directory.
No Format cp hadoop-common-project/hadoop-common/target/site/jdiff/xml/Apache_Hadoop_Common_${version}.xml hadoop-common-project/hadoop-common/dev-support/jdiff cp hadoop-hdfs-project/hadoop-hdfs/target/site/jdiff/xml/Apache_Hadoop_HDFS_${version}.xml hadoop-hdfs-project/hadoop-hdfs/dev-support/jdiff find hadoop-yarn-project -name "Apache_Hadoop_YARN_*_${version}.xml" | xargs -I{} cp {} hadoop-yarn-project/hadoop-yarn/dev-support/jdiff find hadoop-mapreduce-project -name "Apache_Hadoop_MapReduce_*_${version}.xml" | xargs -I{} cp {} hadoop-mapreduce-project/dev-support/jdiff
Update
hadoop-project-dist/pom.xml
No Format <jdiff.stable.api>X.Y.Z</jdiff.stable.api>
- In Nexus
- effect the release of artifacts by selecting the staged repository and then clicking
Release
- If there were multiple RCs, simply drop the staging repositories corresponding to failed RCs.
- effect the release of artifacts by selecting the staged repository and then clicking
- Wait 24 hours for release to propagate to mirrors.
- Edit the website (Generic docs about the new website generation can be found [here|How+to+generate+and+push+ASF+web+site)
Checkout the website if you haven't already
No Format git clone https://gitbox.apache.org/repos/asf/hadoop-site.git -b asf-site
- Install hugo if you haven't already ((tldr; apt-get install/pacman -S/brew install hugo))
Create the new release announcement (Set environment variable version first.
export VERSION=X.Y.Z).
No Format cat << EOF > src/release/${VERSION}.md --- title: Release ${VERSION} available date: 202X-XX-XX linked: true --- <!--- Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. See accompanying LICENSE file. --> This is the first stable release of Apache Hadoop TODO line. It contains TODO bug fixes, improvements and enhancements since TODO. Users are encouraged to read the [overview of major changes][1] since TODO. For details of TODO bug fixes, improvements, and other enhancements since the previous TODO release, please check [release notes][2] and [changelog][3]. [1]: /docs/r${VERSION}/index.html [2]: http://hadoop.apache.org/docs/r${VERSION}/hadoop-project-dist/hadoop-common/release/${VERSION}/RELEASENOTES.${VERSION}.html [3]: http://hadoop.apache.org/docs/r${VERSION}/hadoop-project-dist/hadoop-common/release/${VERSION}/CHANGELOG.${VERSION}.html EOF
- Note: update all the TODO + the date. Don't use date from the future, it won't be rendered.
- Remove the
linked: true
line from the previous release file, eg. from src/release/3.0.0.md. Docs/downloads of the releases withlinked:true
will be linked from the menu. add the docs and update the
content/docs/current
link, by doing the following:No Format cd content/docs tar xvf /path/to/hadoop-${version}-site.tar.gz # Update current2, current, stable and stable2 as needed. # For example rm current2 current ln -s r${version} current2 ln -s current2 current
- Similarly update the symlinks for stable if need be.
- Check the rendering of the new site:
hugo serve && firefox http://localhost:1313
Regenerate the site, review it, then commit it per the instructions in HowToCommit. (The generated HTML files also should be committed. Both src and the rendered site are in the same repo.)
No Format hugo git add . git commit git push
- Send announcements to announce@apache.org and to the user and developer lists once the site changes are visible. Optionally on Hadoop-PMC owned twitter account @hadoop. (For password reach out to private@)
- In JIRA, close issues resolved in the release. Disable mail notifications for this bulk change. Recommend not closing, since it prevents JIRAs from being edited and makes it more difficult to track backports.
- Add the release in "Apache Committee Report Helper" for the next board report to pick that up automatically at https://reporter.apache.org/addrelease.html?hadoop (Requires PMC member rights)
- Update the DOAP file with the release details, DOAP File. The changes will be reflected here after almost 1 hour, if done correctly.
Docker images
Docker images containing release binaries are accessible at DockerHub. A docker image is automatically built by pushing a commit to one of the two special branches: docker-hadoop-2 and docker-hadoop-3. An example is HADOOP-18681. Please contact Ayush Saxena and Wei-Chiu Chuang who have admin privilege to the Apache Hadoop DockerHub repo and can add additional tags to the docker image produced.
...