The load balancing architecture

Before we dive into the Puppet module world again, let's have a quick look at the system architecture and how the load balancer fits into it. The load balancer will be installed on a dedicated virtual machine, which sits in front of the web and the monitoring servers.

First, we will configure the load balancer in the so-called proxy mode, which means that the load balancer is acting like a router for HTTP requests that it forwards to the appropriate backend server based on the URL that the user is accessing.

For example, when a user requests resource http://loadbalancer_ipaddress/nagios, the load balancer forwards the request to the monitoring server, and when the requested resource is http://loadbalancer_ipaddress/webapp, the request is routed to the web server.

Later, we will configure the load balancer, do some real load balancing between two web server nodes and experience how load balancer makes services more resilient against outages when one of the web...