Project Description

Develop a tool to collect runtime information about a kubernetes cluster.

This will aide in the debugging process by capturing cluster events and logs in a way that easy to analyze, where otherwise you would have to manually watch these resources via console. Ideally, we will also be able to apply filters to the collection to narrow the scope of gathered resources.

This will be especially useful when a resource has a short lifetime and get cleaned up before the user can manually inspect it, such as a failed pod for a job.

I started the project a few weeks ago and have made some progress already. It currently can pull some common resources (pods, jobs, services, etc) and dump them to the filesystem as yaml files.

Goal for this Hackweek

There are a few big features I would like to see finished:

1. Internalize the current collector into the cluster
2. collect pod logs

Resources

https://github.com/joshmeranda/kubedump