You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 98 Next »

{{{#!wiki red/solid IMPORTANT NOTE: Apache has limited QA capabilities. It is extremely important that everyone avoids submitting patches to Jenkins for testing unless they are absolutely certain that all relevant tests will pass. HDFS unit tests, for example, will tie up a test slot for over two hours.
}}}

Information

With over 900 patches not yet reviewed and approved for Apache Hadoop, it's time to make some strong progress on the bug list!

A number of Apache Hadoop committers and Hadoop-related tech companies are hosting an Apache Hadoop Community event on Friday, May 8th, after HBaseCon 2015. You are hereby invited for a fun, daylong event devoted to identifying and registering important patches and cleaning up the queue.

How to Get Involved

For this first event, we are primarily targeting individuals with some experience either working with Hadoop or working with large, open source projects. In order to hit the ground running on that day, it is recommended some pre-work be done first:

  1. Have or create an account on Apache JIRA.
  2. Be familiar with the instructions on HowToContribute.
  3. Have a working development and test environment. See BUILDING.txt for requirements as well as how to execute a script to create a Docker image that has all of the components.
  4. Register with https://www.eventbrite.com/e/apache-hadoop-global-bug-bash-tickets-16507188445
  5. Hang out on the #hadoop channel on irc.freenode.net .

Procedures for that day

Source Code Contributions

Non-committers

  1. Read through the To Be Reviewed queue.
  2. Find a JIRA to work on.
  3. Remove the BB2015-05-TBR label from that JIRA so that it leaves the queue.
  4. Work through the issues in that JIRA:
    1. Name the patch file (something).patch
    2. Verify the patch applies cleanly
    3. Fix any pre-existing comments

    4. Test the patch locally using test-patch.sh [See TestPatchTips for more!]

  5. Before uploading, did you run test-patch.sh?
  6. Upload the reworked patch back into JIRA.
  7. Set the label to BB2015-05-RFC.

Note that in some cases, the functionality in older patches may already exist. Please close these JIRA, preferably as a duplicate to the JIRA that added that functionality or as Invalid with a comment stating that you believe the issue is stale and already fixed.

There is also a possibility that the patch requires no further changes and is ready for the committer to review. In that case, just change the label.

Committers

  1. Read through the Ready For Committer queue.
  2. Find a JIRA to work on.
  3. Remove the BB2015-05-RFC label from that JIRA so that it leaves the queue.
  4. Review the patch. If it needs more work, add a comment and add the* BB2015-05-TBR* label so that it goes back into the non-committer queue.
  5. Commit the patch as per usual if it is ready to go.

Non-source Code Contributions

The vast majority of Hadoop content, including almost all of the documentation is part of the source tree. However, there are multiple ways in which those who are unfamiliar with developer environments can contribute:

  1. Go through the open JIRA list and verify whether it is functionality that already exists, a bug that has already been fixed, or is still a valid issue.
  2. Create a wiki account and update the documentation here.
  3. Collaborate with another person in testing, documentation, and other non-source related tasks that almost all patches require.
  • No labels