mirror of https://github.com/apache/kafka.git
Previously, Apache Kafka was uploading release candidate (RC) artifacts to users' home directories on home.apache.org. However, since this resource has been decommissioned, we need to follow the standard approach of putting release candidate artifacts into the appropriate subversion directory, at https://dist.apache.org/repos/dist/dev/kafka/. Reviewers: Justine Olshan <jolshan@confluent.io> |
||
---|---|---|
.. | ||
README.md | ||
git.py | ||
gpg.py | ||
notes.py | ||
preferences.py | ||
release.py | ||
requirements.txt | ||
runtime.py | ||
svn.py | ||
templates.py | ||
textfiles.py |
README.md
Releasing Apache Kafka
This directory contains the tools used to publish a release.
Requirements
- python 3.12
- git
- gpg 2.4
The full instructions for producing a release are available in https://cwiki.apache.org/confluence/display/KAFKA/Release+Process.
Setup
Create a virtualenv for python, activate it and install dependencies:
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
Usage
To start a release, first activate the virutalenv, and then run the release script.
source .venv/bin/activate
You'll need to setup PUSH_REMOTE_NAME
to refer to
the git remote for apache/kafka
.
export PUSH_REMOTE_NAME=<value>
It should be the value shown with this command:
git remote -v | grep -w 'github.com' | grep -w 'apache/kafka' | grep -w '(push)' | awk '{print $1}'
Then start the release script:
python release.py
Should you encounter some problem, where re-running the script doesn't work, look at the following steps:
- The script remembers data inputted previously if you need to correct it, it is saved under the
.release-settings.json
file in therelease
folder. - If the script is interrupted you might need to manually delete the tag named after the release candidate name and branch named after the release version.