Setup for reverse proxy

Proxying Traffic to a Group of Servers

Follow the steps below

1. Setup an upstream block within the http block

2. nginx configuration

   http

   {
       upstream my_glassfish_cluster 

       {
           server backend1.example.com weight=5;
           server 192.168.99.100:28080;
           server 192.168.99.100:28081 backup;
       }
   }

   Use weight to specify importance (1:lowest - 5:highest), use backup to specify only route to this server if all other servers in the cluster fail

3. Now specify the upstream group that nginx will pass requests to (can be any one of the following protocols: proxy_pass, fastcgi_pass, memcached_pass, uwsgi_pass, or scgi_pass)

4. nginx configuration

   http

   {

      ...

      server 

      {
         location / 

         {
            proxy_pass http://my_glassfish_cluster;
         }

      }
   }

   This will pass all http requests up to my_glassfish_cluster server group

Choosing a Load Balancing Method

NGINX supports four load balancing methods in the open source version and one extra method in the paid version

Open source methods

• round-robin (default) - iterate equally around each server in a server group
• least_conn - send requests to server with the least number of connections
• ip_hash - using a hash algorithm, select server in a server group based on client's IP address
• hash -  the server to which a request is sent is determined from a user-defined key which may be a text, variable, or their combination

Paid version

• All of above
• least_time - for each request, NGINX Plus selects the server with the lowest average latency and the least number of active connections, where the lowest average latency is calculated based on which of the following parameters is included on the least_timedirective:
  • header – Time to receive the first byte from the server
  • last_byte – Time to receive the full response from the server