Nginx向应用服务器发送请求的方式有哪些？求含proxy_pass等的完整列表

Great question! Nginx has a solid set of built-in ways to route traffic to your application servers, tailored to different protocols your app might speak. Let’s break down all the main supported methods clearly:

• HTTP/HTTPS Proxying (proxy_pass)
  This is the most widely used method for forwarding requests to HTTP-based apps (like Node.js services, Python Flask/Django apps running an HTTP server, Java Spring Boot apps, etc.). You define a location block and use proxy_pass to point to your app's HTTP endpoint—this can be a local port, remote IP, or domain. Example config:

  location / {
      proxy_pass http://localhost:3000;
      proxy_set_header Host $host;
      proxy_set_header X-Real-IP $remote_addr;
  }
  

• FastCGI Pass (fastcgi_pass)
  Built for apps that use the FastCGI protocol, most commonly PHP applications (WordPress, Laravel, etc.). You’ll pair this with fastcgi_param to pass critical environment variables to services like PHP-FPM. Example:

  location ~ \.php$ {
      fastcgi_pass unix:/run/php/php8.2-fpm.sock; # Or tcp://localhost:9000 for TCP connections
      fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
      include fastcgi_params;
  }
  

• uWSGI Pass (uwsgi_pass)
  Designed specifically for Python apps using the uWSGI protocol (a common setup for Django/Flask deployed with uWSGI servers). Nginx can communicate with uWSGI via TCP or Unix sockets. Example:

  location / {
      uwsgi_pass unix:/path/to/your/app.sock; # Or tcp://localhost:8000
      include uwsgi_params;
  }
  

• SCGI Pass (scgi_pass)
  A lighter alternative to FastCGI, using the SCGI protocol. It’s used for legacy or lightweight apps (some Perl/Python services) that implement SCGI. Example:

  location /scgi-endpoint {
      scgi_pass localhost:9000;
      include scgi_params;
  }
  

• gRPC Proxying (grpc_pass)
  For apps built with gRPC (a high-performance RPC framework), Nginx supports forwarding gRPC requests over HTTP/2. You’ll need to enable HTTP/2 in your server block first. Example:

  server {
      listen 443 ssl http2;
      ssl_certificate /path/to/ssl/cert.pem;
      ssl_certificate_key /path/to/ssl/key.pem;
  
      location /mygrpc.Service/ {
          grpc_pass grpc://localhost:50051;
      }
  }
  

• WebSocket Proxying
  While there’s no dedicated "websocket_pass" directive, Nginx can proxy WebSocket connections by upgrading the HTTP connection. Use proxy_pass alongside specific headers to handle the handshake:

  location /ws {
      proxy_pass http://localhost:8080;
      proxy_http_version 1.1;
      proxy_set_header Upgrade $http_upgrade;
      proxy_set_header Connection "upgrade";
  }
  

• Memcached Proxying (memcached_pass)
  If your app uses Memcached for caching, Nginx can directly forward requests to the Memcached server to serve cached content without hitting your app server. You can add a fallback to your app if the cache misses:

  location /cached-content {
      set $memcached_key $uri;
      memcached_pass localhost:11211;
      error_page 404 = @fallback_to_app;
  }
  
  location @fallback_to_app {
      proxy_pass http://localhost:3000;
  }
  

These are all the primary built-in methods Nginx provides to connect to application servers or backend services. The right choice depends entirely on the protocol your application is configured to use.

内容的提问来源于stack exchange，提问作者Pablo Tato Ramos