New module for managing hadoop clusters with Apache Ambari

Archives Ansible Developer
1

We recently released a custom module that we use to manage our hadoop infrastructure with Ambari. There isn’t anything else out there like this, and it has really simplified our codebase. I would love to work with the community to contribute this into Ansible as an extra module if there is demand. Rather than blindly submitting a PR, I figured I would gauge interest. If there is interest, what would be the best way to move forward?

The code is here (along with an example role):
https://github.com/mbittmann/ambari-ansible-module

Here is a blog post with some context and background:
https://medium.com/@b23llc/managing-distributed-data-products-with-ansible-and-ambari-44c7175555d8

2

actually submitting a PR is the best way, most users are not on ansible-devel

3

Oh cool thanks. Will do. Any guidance on which package under ansible-modules-extras? I see an elasticsearch module under ‘packaging’, and some nosql modules under ‘database/misc’. My gut says the best place for an Ambari hadoop cluster management module would be in the ‘clustering,’ but it doesn’t quite fit with consul. I wonder if there should be a ‘big_data’ and/or ‘nosql’ package?