GnoseedGnoseed
DevOps · 6 modules

Kubernetes Troubleshooting Scenarios

Practice Kubernetes the way you hit it in production — as scenarios. Diagnose real failures across architecture, workloads, networking, storage, configuration and security, and remember the fixes with spaced repetition.

Start learning See sample questions
flashcards
80
flashcards
per day
~10 min
per day
level
Intermediate → Advanced
level
modules
6
modules
About this topic

Learn Kubernetes by debugging it

Knowing the Kubernetes primitives is one thing; diagnosing why a rollout is throwing 503s, a PVC is stuck Pending, or a Service returns no endpoints is another. This track is built entirely from realistic scenarios — a symptom, some evidence, and the question of what is actually going wrong.

Each card drops you into a concrete production situation across cluster architecture, workloads, networking, storage, configuration and security, then walks through the root cause and the fix. It is the difference between knowing the docs and knowing what to do at 2 a.m.

Spaced repetition resurfaces these patterns so the next time you see the symptom, the cause comes to mind fast.

What you'll learn

6 modules, seed to bloom

Each module is a set of flashcards — 80 in total. Answer, review, and watch your knowledge grow from seed to full bloom.

Cluster Architecture (Advanced)

Scenario-based questions on control plane, nodes, and core components

15 cards

Workloads & Pods (Advanced)

Scenario-based questions on Pods, Deployments, StatefulSets, DaemonSets, Jobs

5 cards

Networking & Services (Advanced)

Scenario-based questions on Services, Ingress, DNS, and traffic routing

15 cards

Storage (Advanced)

Scenario-based questions on Volumes, PVs, PVCs, StorageClass, and CSI

15 cards

Configuration (Advanced)

Scenario-based questions on ConfigMaps, Secrets, resource management, and QoS

15 cards

Security (Advanced)

Scenario-based questions on RBAC, ServiceAccounts, NetworkPolicy, and Pod security

15 cards
Try before you plant

Sample questions

A taste of the real flashcards. Pick an answer, then reveal the explanation.

Sample · Kubernetes Troubleshooting Scenarios

A developer creates a Service targeting Pods with label 'app: backend', but the Service returns no endpoints. Running 'kubectl get endpoints my-service' shows an empty list. The Pods are running and healthy. What is the most likely cause?

  • AThe Service selector does not match the Pods' labels: one of them is misspelled, and selectors match exactly
  • BThe Service sits in a different namespace, and cross-namespace routing needs an opt-in annotation on the Pods
  • CThe Pods are running but not Ready, and a Service ignores Pods for their first 30 seconds
  • Dkube-proxy has not synced yet, and a new Service has a mandatory five-minute propagation delay
Permalink & share
Sample · Kubernetes Troubleshooting Scenarios

A developer creates a PersistentVolumeClaim requesting 10Gi of storage with StorageClass 'fast-ssd'. The PVC stays in Pending state. Running 'kubectl describe pvc' shows: 'waiting for first consumer to be created before binding'. What does this mean?

  • AThe StorageClass uses volumeBindingMode: WaitForFirstConsumer, so nothing is provisioned until a Pod actually needs the claim
  • BThe cluster has run out of SSD storage capacity, and the PVC is waiting in a queue until storage is freed by other workloads being terminated
  • CAn administrator must approve the request by hand, because that StorageClass requires approval above 5Gi
  • DThe CSI driver for the 'fast-ssd' StorageClass has crashed and the PVC is retrying the provisioning request every 30 seconds until the driver recovers
Permalink & share
How Gnoseed works

Learn it once, keep it for good

1

Answer a question

Each card is one practical concept with multiple options. Pick what you think is right.

2

Get the full answer

See the correct option plus a clear explanation, and a link to deeper docs when one is available.

3

Review at the right time

A spaced-repetition engine (SM-2 or FSRS) resurfaces each card just before you would forget it.

Why learn this

Why scenario practice works

Train the real skill

Production work is diagnosis, not recall. These cards practice the actual thing you get paid for.

Pattern-match faster

Seeing many failures means the next outage looks familiar instead of foreign.

Interview-ready

"Walk me through how you would debug this" is a staple of senior Kubernetes interviews.

FAQ

Common questions

Is this for beginners? +

It assumes you know the Kubernetes basics. If the primitives are new, start with the Kubernetes Fundamentals track first, then come here.

How long does it take? +

About 10 minutes a day. Scenario questions take a little longer to read, so the timing adapts — short, frequent sessions still win.

Does this prep me for the CKA? +

Gnoseed is a study companion, not a course. These scenarios map closely to the CKA troubleshooting domain, so they pair well with hands-on practice to make the diagnoses stick.

Is it free? +

Yes, completely free. No registration or credit card is required, and all your progress is stored locally in your browser.

Ready to debug Kubernetes with confidence?

Plant your first seed today. Ten minutes a day is all it takes to grow real production troubleshooting instincts.

Start learning free