NGNIX - Load Balancing

The upstream directive

The upstream directive defines groups of servers that can be referenced by the proxy_pass, fastcgi_pass, uwsgi_pass, scgi_pass, memcached_pass, and grpc_pass directives:

upstream testdomainapp  {
   server app1.testdomain.local:8085;
   server app2.testdomain.local:8086 backup;
   server app3.testdomain.local:8087 backup;
}

By default, requests are distributed between the servers using a weighted round-robin balancing method.

The location directive

The location directive is the other part of the equation that enables the upstream directive:

location /app {
        proxy_pass http://testdomainapp/;
}

In this example, we are using the /app location to front-end the testdomainapp server group, defined in the upstream directive.

The server directive

The server directive defines an individual server that is part of the upstream group:

upstream testdomainapp  {
   server app1.testdomain.local:8085 weight=3;
   server app2.testdomain.local:8086;
   server app3.testdomain.local:8087 backup;
}

There are may options available to define how each server is handled in the group. Server parameters available in the non-commercial NGINX offering include:

• weight= - Sets the weight of the server, default=1
• max_conns= - Limits active connections, default=0
• max_fails= - Maximum unsuccessful connections, default=1
• fail_timeout= - Sets fail timeout, default=10 seconds
• backup - Marks the server as a backup server
• down - Marks the server as a unavailable