Nginx完全指南 (Derek DeJonghe) (Z-Library)

【One-Line Pitch】 A hands-on recipe collection for engineers who already understand modern web architecture and now need to configure NGINX (open source or Plus) for real application-delivery problems—load balancing, security, caching, cloud/container deployment, and operations. Best for practitioners who want working configurations and the reasoning behind them, not a from-zero tutorial. 【Book Arc】 - **Opening (~0%–13%)**: Installation and fundamentals—setting up NGINX on Debian/Ubuntu and RedHat/CentOS, installing NGINX Plus, verifying the install, and learning the key files, directories, and commands. Solves the "get it running and know where things live" problem. - **Early (~13%–33%)**: Core delivery mechanics—HTTP/TCP/UDP load balancing, session persistence, health checks, traffic management (A/B testing, GeoIP, rate/connection/bandwidth limits), large-scale caching, and programmability/automation (Plus API, key-value store, njs, Chef/Ansible/Consul). - **Middle (~33%–53%)**: Security and modern protocols—authentication (basic auth, subrequests, JWT, OpenID Connect), access control and TLS, HTTP/2 and gRPC, plus media streaming (MP4/FLV, HLS/HDS). - **Late (~53%–67%)**: Deployment environments—cloud (AWS, Azure, Google Compute Engine), containers and microservices (API gateway, Docker images, Kubernetes Ingress, service mesh/mTLS), high-availability patterns, and monitoring. - **Ending (~67%–100%)**: Operations and tooling—logging and request tracing, performance tuning, NGINX Instance Manager and Controller, and practical ops tips like includes and config debugging. 【Key Takeaways】 - **This is a cookbook, not a linear textbook** (Early): each section is framed as problem → solution → explanation, so it works best as a reference you jump into by task. - **Load balancing spans HTTP, TCP, and UDP** (Early): the `upstream` pattern recurs across the `http` and `stream` contexts, with algorithms like round-robin, least connections, and Plus-only least-time. - **Session persistence is a first-class concern** (Early): NGINX Plus offers sticky cookie, sticky learn, and sticky routing for stateful applications where requests must return to the same server. - **Health checks split into passive (open source) and active (Plus)** (Early): passive checks watch live traffic, while active checks probe upstreams proactively to avoid client-facing failures. - **NGINX is often the first security boundary** (Middle): the book treats authentication, TLS, access control, and WAF/DDoS protection as core delivery concerns, not add-ons. - **Caching and traffic management are tunable levers** (Early): cache zones, locking, hash keys, bypass, and slicing, alongside rate, connection, and bandwidth limits. - **Cloud, containers, and service mesh are covered as deployment targets** (Late): AWS/Azure/GCP patterns, Docker images, Kubernetes Ingress, and mTLS mesh setups. - **Operations close the loop** (Ending): logging, tracing, performance tuning, and management tooling (Instance Manager, Controller) round out the lifecycle. 【Reading Tips】 - Use the table of contents as your entry point: jump straight to the chapter matching your current task rather than reading cover to cover. - Deep-read the load balancing, security, and caching chapters—these are the highest-leverage for production delivery work. - Skim the installation and cloud-provider chapters if you already have a working environment; return when migrating platforms. - Watch for the recurring "NGINX Plus only" annotations—many advanced features (active health checks, least-time, JWT, sticky sessions) are not in the open source edition. - Treat the config snippets as starting points and always validate with `nginx -t` before reloading. 【Coverage Limits】 This guide is synthesized from stratified excerpts covering the front matter, table of contents, and early-to-mid chapters; later chapters on Instance Manager, Controller, and detailed ops tips are only visible at the heading level, so their specific content is not summarized here.