iNTERFACEWARE™ Inc.

What is a Failover Cluster?

A failover cluster typically consists of two redundant nodes, but sometimes there are more than two, depending on the complexity of the setup. The primary node (or server) is used to provide the service in a normal operating state, while the secondary server is used to run the service should the primary server fail.

Typical Failover Cluster Setup

The primary and secondary servers should share some common characteristics. For example:

• Each server should have Iguana running with identical configurations.
• Each server should have access to the same external database. This database should have built-in redundant strategies.
• Both servers should share the same IP address. If a failure occurs, the failed system will release the active IP address, and the secondary server taking over will use the same IP address. It will then continue listening for inbound connections.

In a normal situation the primary server is monitored every few seconds to ensure that it is still active and functioning properly. The system is typically monitored by a secondary system, which could either be a software- or hardware-enabled solution. In some cases, the system performing the monitoring could even be the secondary failover server. If Iguana should stop running properly on the primary server or a hardware failure occurs, the monitoring system will detect the failure and the secondary node will take over.

The main goal of failover clustering is to provide an effective means of continuing the Iguana service should the primary server fail due to a software or hardware component failure. By having a clustering strategy in place, you can ensure that an exact instance of Iguana is available to your customers while the other system is being fixed.