Capstone: Follow the Request

This final section deliberately ties every module in this track together by tracing one single request all the way through a realistic system, end to end. Starting from outside the cluster: an external gateway or load balancer receives the request; an Ingress or Gateway route (from the networking module) matches it by host and path; that route points at a Service, which resolves to a specific ready Pod via its EndpointSlice (also networking); a NetworkPolicy (from its own module) must permit that connection to actually reach the Pod; the Pod itself reads its configuration and Secrets (configuration module) and authenticates as its ServiceAccount (security module); it then calls its own downstream dependencies through their own Services; and any durable writes land in correctly-provisioned, correctly-attached persistent storage (storage module). Once you can trace that full static chain confidently, overlay the dynamic behaviors from the rest of the track on top of it: what does a rolling update (Deployments module) do to this chain mid-request, how does autoscaling (resources module) change capacity along the way, and how does planned node maintenance (cluster-ops module) affect it? Holding this entire, connected picture in mind at once is what separates isolated command knowledge from genuine production judgment - and it is precisely why the CKAD-style application skills and the CKA-style cluster-operator skills taught throughout this track were never really two separate subjects, only two different vantage points on the same request path.

Analogy: Kubernetes is a thermostat, not a remote control. You declare the temperature; independent controllers measure reality and keep acting. Debugging means finding which sensor, rule or actuator prevents convergence.

Read the object as evidence

kubectl get RESOURCE NAME -n NAMESPACE -o yaml
kubectl describe RESOURCE NAME -n NAMESPACE
kubectl get events -n NAMESPACE --sort-by=.metadata.creationTimestamp
kubectl explain RESOURCE.spec

Do not memorize these as a ritual. The first command exposes desired and observed state, the second connects conditions and events, the third supplies a timeline, and the fourth checks the server's schema. Compare what the controller was asked to do with what it reports doing. Then test the narrowest hypothesis.

Scenario: The platform's own linked Kong, Redis, Postgres, PgBouncer, and Kafka labs are deliberately designed to cross every boundary this capstone describes, rather than staying neatly inside one module's topic. Treat the full set of linked labs as one connected production incident to reason through end to end - drawing an actual request-path diagram, forming an explicit failure hypothesis, gathering an evidence table exactly as the observability module taught, and finally verifying the recovery - rather than as twenty separate, disconnected command exercises to clear one at a time.

Follow one request end to end

Start outside the cluster: DNS and load balancer reach a gateway listener. Host and path select a route; the backend reference selects a Service port; EndpointSlice supplies ready Pod IPs; NetworkPolicy permits the connection; the Pod reads configuration and identity; dependencies resolve through Services; durable writes reach correctly attached storage. Overlay rollout, autoscaling and maintenance at every hop.

Scenario: Kong returns 503 after an HPA event. The route and Service are unchanged, but new Pods fail readiness because a rotated database Secret was consumed only by old environment state. Cross-layer timelines expose the chain.
Goal: Run the linked Kong, Redis, Postgres, PgBouncer and Kafka exercises as one incident. Write a request-path diagram, failure hypothesis, evidence table and recovery verification suitable for a production handoff.

Production reasoning

Ask four questions: Who owns this object? What dependency must become ready next? Which controller reports the blocking condition? What evidence would disprove my current theory? This prevents symptom-driven changes. Record the context, namespace, object generation, image digest and recent rollout before mutation; a recreated Pod may erase the evidence you needed.

Warning: Running is not the same as ready, healthy, durable or correct. Kubernetes status is layered. Confirm the application-level outcome as well as the object state.
Goal: Put this model into practice in the Kubernetes lab networkpolicy-kong-routing. Open /labs/kubernetes, choose networkpolicy-kong-routing, predict the failure path before changing anything, then use the simulator's check command to validate the finished state.

Deliberate practice

Before the lab, write the expected object relationship and the first three commands you will run. Afterward, explain why the fix converged and name one tempting change that would only mask the symptom. Repeat using an explicit namespace and a structured output format. This prediction-observation-explanation loop is what turns command familiarity into production judgment.

45-minute investigation

  1. Map (5 min): draw the owner-to-child chain and mark every namespace, selector, identity and dependency involved.
  2. Predict (5 min): write one expected status condition, one likely event and one log or metric signal before opening the lab.
  3. Observe (10 min): collect YAML, describe output and ordered events. Do not mutate state. Record which observation disproves your first theory.
  4. Repair (15 min): make the smallest declarative correction, watch the responsible controller converge, and verify the user-facing path rather than stopping at Running.
  5. Stress (10 min): change one relevant constraint - replica count, label, readiness, resource value or placement rule - predict the outcome, observe it, then restore the known-good declaration.
Tip: Keep a short incident note with symptom, evidence, hypothesis, change, result. Across four sections this produces a reusable runbook instead of a pile of remembered commands.