普通配置

upstream www {
  server 172.17.0.2;
  server 172.17.0.3;
}


server {
  listen 80;
  listen [::]:80;
  server_name 172.17.0.4;
  client_max_body_size 1024m;
  location / {
    proxy_pass http://www;
    proxy_set_header HOST $host;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  }
}

随机分配后端服务器

自定义权重

upstream www {
  server 172.17.0.2 weight=6;
  server 172.17.0.3 weight=3;
}


server {
  listen 80;
  listen [::]:80;
  server_name 172.17.0.4;
  client_max_body_size 1024m;
  location / {
    proxy_pass http://www;
    proxy_set_header HOST $host;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  }
}

权重越大，被访问的次数就越多

权重+哈希

upstream www {
  ip_hash;
  server 172.17.0.2 weight=6;
  server 172.17.0.3 weight=3;
}


server {
  listen 80;
  listen [::]:80;
  server_name 172.17.0.4;
  client_max_body_size 1024m;
  location / {
    proxy_pass http://www;
    proxy_set_header HOST $host;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  }
}

在 upstream 指令块中增加了 ip_hash 指令。该指令就是告诉 nginx 服务器，同一个 IP 地址客户端发送的请求都将分发到同一个 后端服务器进行处理,此类一般用于有登录操作的场景

服务器状态负载

upstream www {
  server 172.17.0.2 weight=6;
  server 172.17.0.3 weight=3;
  fair;
}


server {
  listen 80;
  listen [::]:80;
  server_name 172.17.0.4;
  client_max_body_size 1024m;
  location / {
    proxy_pass http://www;
    proxy_set_header HOST $host;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
  }
}

根据服务器处理请求的时间来进行负载，处理请求越快，也就是响应时间越短的优先分配