Kubernetes Architecture & Production Operations

A complete, practical study guide for intermediate-to-senior platform engineers. This module explains the Kubernetes control plane, the lifecycle of a resource from API request to running workload, distributed reconciliation, placement, stateful workloads, scaling, security and production operations.

Scope and safety

All marketplace and infrastructure examples are hypothetical. Exact behavior can vary by Kubernetes version, add-on, CNI/CSI implementation and managed service. Validate commands before applying them in an authorized environment. Never publish credentials, private cluster configuration or production logs.

Prerequisites

• Familiarity with Linux processes, TCP/IP, containers and basic YAML.
• Basic kubectl, Deployments and Services experience.
• Access to an isolated Kubernetes cluster for hands-on labs.

Learning objectives

By the end, you should be able to trace a Deployment through admission, etcd, controllers, scheduling and kubelet; explain Raft quorum and watch recovery; implement safe reconciliation; diagnose placement, resource, probe and volume failures; design autoscaling and access controls; and upgrade a production cluster with verified compatibility and recovery boundaries.

Chapters

1. Architecture: control plane, nodes and failure boundaries
2. Complete Deployment lifecycle: apply to serving traffic
3. etcd, Raft, quorum, watches and recoverability
4. Controllers, informers, ownership and finalizers
5. Scheduler filtering, scoring and topology
6. Deployments, StatefulSets, DaemonSets, Jobs and CronJobs
7. Resources, QoS, throttling, eviction and priority
8. Probes, graceful termination, readiness gates and PDBs
9. Storage: PV, PVC, CSI, topology and snapshots
10. Autoscaling with HPA, VPA, KEDA and node provisioning
11. Security: authentication, RBAC, admission and isolation
12. Upgrades, compatibility and production troubleshooting

Practice

• Six hands-on labs with commands, expected behavior and safety notes
• 42 advanced technical FAQs

Recommended learning sequence

Start with Chapters 1–4 to establish the control-plane model. Then study placement and workload guarantees in Chapters 5–6; use Chapters 7–10 to reason about resource, storage and scale failures. Finish with security and upgrades, and revisit the 42-question challenge bank without looking at answers. Draw the sequence diagrams and reproduce the labs in an isolated environment.

Primary technical references

• Kubernetes architecture
• Admission controllers
• Scheduler framework
• Workload controllers
• etcd documentation
• Pod lifecycle
• Storage classes
• HPA internals
• RBAC
• Version skew policy

← All guides · Start Chapter 1 →