devops/activities/06-kubernetes-intro-homework.md
2026-01-01 00:00:00 +00:00

60 lines
2.3 KiB
Markdown

---
id: act-06-kubernetes-intro-homework
title: Kubernetes Intro - Homework
mode: homework
open_at: 2026-02-01T00:00:00+01:00
close_at: 2026-12-31T23:59:59+01:00
retakes_enabled: true
max_attempts: 999
grade_max: 100
---
## T1
Type: essay
Points: 30
Prompt: Compare control plane nodes and worker nodes by listing the main responsibilities and naming at least three specific components/services on each side.
Refs:
- mat-06-kubernetes-intro#control-plane-and-worker-nodes
- mat-06-kubernetes-intro#control-plane-services
- mat-06-kubernetes-intro#worker-node-components
Rubric:
- Correctly distinguishes control plane responsibilities vs worker responsibilities. (10)
- Correctly names and attributes at least three control plane components/services (e.g., API server, scheduler, controllers/controller manager, cluster store). (10)
- Correctly names and attributes at least three worker node components (kubelet, runtime, kube-proxy) and what they do. (10)
## T2
Type: mcq
Points: 20
Prompt: Which control plane component is described as the front end that all commands and requests go through?
Refs:
- mat-06-kubernetes-intro#the-api-server
Choices:
- [x] API server
- [ ] kubelet
- [ ] kube-proxy
- [ ] container runtime
## T3
Type: short
Points: 20
Prompt: What distributed database is used as the basis for the Kubernetes cluster store?
Refs:
- mat-06-kubernetes-intro#the-cluster-store-etcd-high-availability-and-split-brain
Answer: etcd
## T4
Type: single_choice
Points: 15
Prompt: If etcd enters a split-brain condition, what behavior is described? A) etcd deletes all cluster state B) etcd goes into read-only mode preventing updates C) etcd automatically adds replicas until quorum is restored D) etcd converts the cluster to single-node mode
Refs:
- mat-06-kubernetes-intro#the-cluster-store-etcd-high-availability-and-split-brain
## T5
Type: multiple_choice
Points: 15
Prompt: Which statements about Pods are consistent with the material? A) Pods are the unit of scheduling B) All containers in a Pod can be scheduled to different nodes C) A Pod is ready only when all its containers are running D) Pods are immutable and updates replace them with new Pods
Refs:
- mat-06-kubernetes-intro#pod-scheduling-atomic-readiness
- mat-06-kubernetes-intro#pods-and-containers
- mat-06-kubernetes-intro#pod-immutability