check_k8s: Kubernetes plugin for Nagios

Nagios plugin for monitoring Kubernetes Clusters, built using the Python standard library.

It currently supports monitoring of Pods, Nodes and Deployments.

Usage

$ python3 check_k8s.py -h

usage: check_k8s.py [-h] [--host HOST] [--token TOKEN]
                    [--token_file TOKEN_FILE] [--port PORT]
                    [--timeout TIMEOUT] [--insecure] [--debug] --resource
                    {pods,nodes,deployments} [--namespace NAMESPACE]
                    [--ignore EXPRESSIONS] [--selector SELECTOR]
                    [--version]

Checks health of a Kubernetes cluster

optional arguments:
  -h, --help            show this help message and exit
  --host HOST           Kubernetes host (default: 127.0.0.1)
  --token TOKEN         Authentication Token
  --token_file TOKEN_FILE
                        Read Token from file
  --port PORT           Kubernetes port (default: 6443)
  --timeout TIMEOUT     Connection timeout in seconds (default: 15.0)
  --insecure            Continue on insecure SSL connection
  --debug               Enable debug mode
  --resource {pods,nodes,deployments}
                        Resource to monitor
  --namespace NAMESPACE
                        Look only within this namespace
  --ignore EXPRESSIONS  Regular Expression to match against the resource
                        names to ignore in the check results. Can be
                        invoked multiple times.
  --selector SELECTOR   Label selector query to be used.
  --version             show program's version number and exit

Component system

The check_k8s plugin comes with a simple component system, making it easy to add support for additional Kubernetes Resources.

Resources

Typically, components make use of the k8s.Resource class, which provides common functionality for working with Kubernetes APIs.

Mappings

The k8s.components.mappings are automatically picked up by the CLI as Resource Enums and provides a glue between input data and component entry points.

Extending

Take a look at the existing components to get started, and check out the Kubernetes API reference for details about the various Kubernetes APIs available.