Lint live Kubernetes clusters for misconfigurations and operational hygiene with Popeye

Inspect a live Kubernetes cluster for unhealthy resource settings, missing probes, and other operational smells.

Prerequisites

Kubernetes cluster access, Popeye binary

Installation

Requirements and caveats from upstream:

• Your cluster must run a metrics-server for these to take place!

• Configure node resources.

• node:

Basic usage or getting-started notes:

• Furthermore, if your cluster employs a metric-server, it reports potential resources over/under allocations and attempts to warn you should your cluster run out of capacity.

• NOTE! This will run Popeye in the context namespace if set or like kubectl will use the default namespace

• Run Popeye in the fred namespace

• Source: https://github.com/derailed/popeye

• Extracted from upstream docs: https://raw.githubusercontent.com/derailed/popeye/HEAD/README.md

Documentation

• https://github.com/derailed/popeye

Source

• Agent Skill Exchange