kubeleds

Monitor your kubernetes clusters using NeoPixel LEDS!

Hardware

Raspberry PI (any model that supports GPIO)

Currently using a Raspberry PI 4b 2GB. Preparing a Raspberry PI Zero W

NeoPixel LED Strips, 12mm Diffussed bulbs

Usage

1. cd into project directory.

2. Create a virtual environment.

$ make venv

3. Activate it.

$ source venv/bin/activate

4. Install development dependencies with editable mode to test the CLI.

$ make install

Take kubeleds for a spin

This will basic command line functionality

$ kubeledscli init

This will retrieve node status and set the leds base off of status of each node, testing each status condition (ready, cpu, memory, disk)

$ kubeledscli get_cluster_nodes '[api fqdn]' \
    '[your token]' 'True' 'get_cluster_nodes' \
    'set_leds' 'get_cluster_nodes' 'Ready' \
    'set_leds' 'get_cluster_nodes' 'MemoryPressure' \
    'set_leds' 'get_cluster_nodes' 'DiskPressure' \
    'set_leds' 'get_cluster_nodes' 'PIDPressure' \
    'set_leds' 'get_cluster_nodes' 'Ready' \

The above commands will retrieve cluster node status and set an led frame per status condition, 1 led per node.

GREEN: Node Ready BLUE: CPU Good CYAN: Memory good Purple: Disk Good

Test with Docker

CLI commands can be tested with Docker.

1. Build an image for the CLI.

   Image is tagged with the same name as the cli_command.

$ make docker-image

2. Run the command inside the container.

$ docker-run --rm kubeledscli init

Documentation

1. Install documentation-related dependencies.

$ make docs

2. Serve the docs locally.

$ make serve-docs

Distribution

NOTE

Make sure you have account in PyPI before you try this out.

To publish you CLI to PyPI, run:

$ make distributions

dist directory will be created inside your project directory. Upload it to PyPI using:

$ twine dist/*

Help

For help related to make commands.

$ make help