Examples

Index of all runnable examples in the tutorial. Each example has its own dedicated page and a runnable demo.sh script in the repo.

⏱ variesSection 16

Every tutorial section that introduces a runnable workload has a matching directory under examples/ in the repo, with a narrated walkthrough (README.md), an end-to-end demo script (demo.sh), and the manifests/values/source needed to run it.

This page is the index. Each example below links to a dedicated page rendered from its README, plus a direct link to the source directory on GitHub so you can clone, copy, or fork it on its own.

How to run an example

Each example is self-contained and idempotent. The basic shape is always:

cd examples/NN-name
./demo.sh

demo.sh includes pre-flight checks (Docker Engine reachable, correct minikube profile with its NodePorts published, required tools in PATH, kernel limits sufficient where applicable) and a cleanup trap so re-runs are safe. Every kubectl, helm, and istioctl call in a demo is pinned to its profile’s context, so the demos never change your current kubectl context. If anything fails, the script exits non-zero with a diagnostic dump of relevant Pod state, recent events, and logs.

To run every example and tally pass/fail:

./scripts/test-all-examples.sh

The aggregator does not fail-fast — it runs every example even when earlier ones fail, then reports pass/fail per script at the end.

All examples

The list below is generated from _example_pages/*.md and sorted by tutorial-section order. Each example links to its dedicated page (rendered from the example’s README.md) and to the source directory on GitHub.

§3 driver check

Read the walkthrough →

Source: examples/03-driver-check/

§6 deploy nginx kubectl

Read the walkthrough →

Source: examples/06-deploy-nginx-kubectl/

§7 nodeport service

Read the walkthrough →

Source: examples/07-nodeport-service/

§8 persistent volume

Read the walkthrough →

Source: examples/08-persistent-volume/

§9 deploy nginx helm

Read the walkthrough →

Source: examples/09-deploy-nginx-helm/

§11 istio

Read the walkthrough →

Source: examples/11-istio/

§12 keda http

Read the walkthrough →

Source: examples/12-keda-http/

§12 keda kafka

Read the walkthrough →

Source: examples/12-keda-kafka/

Verification status

The examples are the canonical record of “what’s been tested vs. what’s claimed”. Each example’s verification status against Fedora 44 is tracked in _plans/reconciliation-plan.md (viewable in the repo). The default status for any new claim is unverified; promotion to verified requires a real test run by a human on Fedora 44.

As of project close-out (r17), the count of verified rows stood at 107. Those runs predate the move to Docker Engine and published NodePorts, so treat the count as historical until the examples are re-verified on the current stack.

What’s NOT an example

Some tutorial sections introduce concepts or commands without ever bundling a runnable example:

  • §1 Prerequisites — reads the host state, doesn’t deploy
  • §2 Tooling install — installs binaries, doesn’t deploy
  • §5 Addons + dashboard — uses minikube addons commands; exploratory, no canonical demo
  • §10 Editor/shell/terminal — installs k9s, kubectx, etc.; no Kubernetes workload to demo
  • §13 Alternatives to minikube — comparative prose only
  • §14 FAQ + cleanup recipes — recipes you copy-paste as needed, not a single linear demo
  • §15 Where to go next — pointers, not a runnable thing

Those sections are valuable in the linear tutorial flow but don’t have a corresponding examples/NN-name/ directory.

← Back to §15: Where to go next