klue

Kubernetes troubleshooting that explains why, not just what.

klue connects to your cluster, builds a resource graph around the object you care about, and runs diagnostic rules to surface root causes — from crash-looping pods and stuck deployments to missing ingress backends and cert-manager certificate failures.

Quick start

go install github.com/gabor-boros/klue@latest
# or `brew install --cask gabor-boros/brew/klue`

klue why pod <name> -n <namespace>  # (1)!

1. Replace <name> and <namespace> with your target Pod. See why for other resource kinds and CRDs.

Already installed?

Run klue version to confirm the binary is on your PATH, then jump to why.

See Installation for other install options.

How it works

flowchart LR
  Connect["Connect to cluster"] --> Discover["Discover APIs"]
  Discover --> Fetch["Fetch snapshot"]
  Fetch --> Graph["Build graph"]
  Graph --> Rules["Run rules"]
  Rules --> Output["Render diagnosis"]

1. Connect to a cluster using your kubeconfig (or in-cluster credentials).
2. Discover API resources at runtime, including custom resources (CRDs).
3. Fetch a namespace snapshot and build a resource graph.
4. Run diagnostic rules against the graph and related events.
5. Render a human-readable (or JSON) diagnosis.

Next steps

Topic	Page
Cluster connection	Kubernetes access
CLI commands	Commands
Flags reference	Flags
Diagnose a resource	why
Contribute	Development