本文最后更新于 69 天前，其中的信息可能已经过时，如有错误请发送邮件到wuxianglongblog@163.com

Nginx服务器的HTTP代理服务

网络初始化之listen常见配置

listen 127.0.0.1:8000;
listen 127.0.0.1;
listen 8000;
listen *:8000;
listen localhost:8000;
listen [::]:8000;
listen [fe80::1];
listen unix:/var/run/nginx.sock;

配置实例一：对所有请求实现一般轮询规则的负载均衡

 http {

      upstream live_node {                        # 配置后端服务器组
          #max_fails默认值为1,fail_timeout默认值为10s,max_fails=0表示不做检查
          server 127.0.0.1:8089 weight=1 max_fails=1  fail_timeout=10s;   
          server 127.0.0.1:8088;
          keepalive 32;
          hash $request_uri consistent;
      }

      server {
          listen 80;
          server_name  localhost;
          location / {
              proxy_pass http://live_node;         # 注意：proxy_pass后面的路径不带uri时，其会将location的uri传递给后端主机
              proxy_set_header Host $host;         # 保留客户端的真实信息
              proxy_set_header Host $host:$server_port;
              proxy_set_header X-Real-IP $remote_addr;
              proxy_set_header X-Real-PORT $remote_port;
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

              proxy_redirect off;
              proxy_buffer_size  128k;
              proxy_buffers   32 32k;
              proxy_busy_buffers_size 128k;
          }
      }

      server {                    # 配置虚拟服务器8088
          listen 8088;
          server_name  localhost;
          location / {
              root   /usr/local/nginx/html2;      
              index  index.html index.htm;
          }
      }

      server {                    # 配置虚拟服务器8089
          listen 8089;
          server_name  localhost;
          location / {
              root   /usr/local/nginx/html3;
              index  index.html index.htm;
          }
      }
  }

Nginx-proxy 详解文章链接
- 查看错误日志
  upstream sent too big header while reading response header from upstream
- Nginx-proxy_buffer_size and fastcgi_buffer
- http://wiki.nginx.org/NginxHttpProxyModule
- http://blog.sina.com.cn/s/blog_5dc960cd0100i4mt.html

参数：keepalive connections;

补充：由于短连接消耗前端代理服务器的资源现象严重,因此会将一部分连接定义为长连接以节省资源
FUN：#为每个worker进程保留的空闲的长连接数量
FUN：#定义nginx与后端服务器的保持连接的数量

参数：hash $request_uri consistent;

FUN：#[consistent]; 使用一致性哈希算法, 建议开启此项
FUN：#基于指定的key的hash表来实现对请求的调度，此处的key可以直接文本、变量或二者的组合；
FUN：#将请求分类，同一类请求将发往同一个upstream server；

配置实例二：对所有请求实现加权轮询规则负载均衡

 http {

      upstream live_node {                        # 配置后端服务器组
          server 127.0.0.1:8089 weight=5;         # 这个处理客户端请求会多些 
          server 127.0.0.1:8088 weight=1;         # 默认 weight = 1
      }

      server {
          listen 80;
          server_name  localhost;
          location / {
              #proxy_pass http://new_uri/;         # 注意：proxy_pass后面的路径是一个uri时，其会将location的uri替换为proxy_pass的uri
              proxy_pass http://live_node;         
              proxy_set_header Host $host;         # 保留客户端的真实信息
          }
      }

      server {                    # 配置虚拟服务器8088
          listen 8088;
          server_name  localhost;
          location / {
              root   /usr/local/nginx/html2;      
              index  index.html index.htm;
          }
      }

      server {                    # 配置虚拟服务器8089
          listen 8089;
          server_name  localhost;
          location / {
              root   /usr/local/nginx/html3;
              index  index.html index.htm;
          }
      }
  }

配置实例三：对特定资源实现负载均衡

 http {

      upstream videobackend {                     # 配置后端服务器组视频代理
          server 127.0.0.1:8088;         
          server 127.0.0.1:8089;        
      }

      upstream filebackend {                      # 配置后端服务器组文件代理
          server 127.0.0.1:8888;        
          server 127.0.0.1:8889;        
      }

      server {
          listen 80;
          server_name  localhost;
          location /video/ {
              proxy_pass http://videobackend;      # 视频代理
              proxy_set_header Host $host;
              proxy_set_header X-Real-IP $remote_addr;        #  proxy_set_header field value; 设定发往后端主机的请求报文的请求首部的值
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;         
          }

          location /file/ {
              proxy_pass http://filebackend;       # 文件代理
              proxy_set_header Host $host;
              proxy_set_header X-Real-IP $remote_addr;
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;         
          }
      }

      server {                    # 配置虚拟服务器8088
          listen 8088;
          server_name  localhost;
          location /video {
               alias   /usr/local/nginx/html2;
          }
      }

      server {                    # 配置虚拟服务器8089
          listen 8089;
          server_name  localhost;
          location /video {
               alias   /usr/local/nginx/html3;
          }
      }

      server {                    # 文件虚拟服务器1
          listen 8888;
          server_name  localhost;
          location /file {
              alias   /usr/local/nginx/html4;
          }
      }

      server {                    # 文件虚拟服务器2
          listen 8889;
          server_name  localhost;
          location /file {
              alias   /usr/local/nginx/html5;
          }
      }
  }

访问方式：http://127.0.0.1/video/demo.txt

输出：this is video HTML2 demo2 8088
输出：this is video HTML3 demo3 8089

访问方式：http://127.0.0.1/file/demo.txt

输出：this is file HTML4 demo4 8888
输出：this is file HTML4 demo5 8889

测数文件：demo.txt

echo "this is video HTML2 demo2 8088" > ./html2/demo.txt

配置实例四：不同的域名实现负载均衡

 http {

      upstream frontend {                     # 配置后端服务器组视频代理
          server 127.0.0.1:8088;         
          server 127.0.0.1:8089;        
      }

      upstream backend {                      # 配置后端服务器组文件代理
          server 127.0.0.1:8888;        
          server 127.0.0.1:8889;        
      }

      server {
          listen 80;
          server_name  www.frontend.com;
          location /video/ {
              proxy_pass http://frontend;      # 前台域名代理
              proxy_set_header Host $host;
              proxy_set_header X-Real-IP $remote_addr;
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;         
          }
      }

      server {
          listen 8088;
          server_name  www.backend.com;
          location /video/ {
              proxy_pass http://backend;      # 后台域名代理
              proxy_set_header Host $host;
              proxy_set_header X-Real-IP $remote_addr;
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;         
          }
      }

  }

配置实例五：实现带有URL重写的负载均衡

 http {

      upstream backend {                      # 配置后端服务器组
          server 127.0.0.1:8888;        
          server 127.0.0.1:8889;        
      }

      server {
          listen 80;
          server_name  www.backend.com;
          index  index.html index.htm;
          location /file/ {
              rewrite ^(/file/.*)/media/(.*)\.*$ $1/mp3/$2.mp3 last;        
          }

          location / {
              proxy_pass http://frontend;      # 前台域名代理
              proxy_set_header Host $host;      
          }
      }

  }

客户端请求URL为：http://www.backend.com/file/download/media/1.mp3
[1]：虚拟主机 location /file/ 块将该URL进行重写为:http://www.backend.com/file/download/media/mp3/1.mp3
[2]：新的URL再有 location / 块转发转发到后端的backend服务器组中实现负载均衡

Nginx服务器的HTTP代理服务

发送评论 编辑评论

推荐文章

发送评论编辑评论