Cluster-Unaware Applications

A cluster-unaware application is distinguished by the following features.

• The application does not use the Failover Cluster API. Therefore, it cannot discover information about the cluster environment, interact with cluster objects, detect that it is running in a cluster, or change its behavior between clustered and non-clustered systems.
• If the application is managed as a cluster resource, it is managed as a Generic Application resource type or Generic Service resource type. These resource types provide very basic routines for failure detection and application shutdown. Therefore, a cluster-unaware application might not be able to perform the initialization and cleanup tasks needed for it to be consistently available in the cluster.

Most older applications are cluster-unaware. However, a cluster-unaware application can be made cluster-aware by creating resource types to manage the application. A custom resource type provides the initialization, cleanup, and management routines specific to the needs of the application.

There is nothing inherently wrong with cluster-unaware applications. As long as they are functioning and highly available to cluster resources when managed as Generic Applications or Generic Services, there is no need to make them cluster-aware. However, if an application does not start, stop, or failover consistently when managed by the generic types, it should be made cluster-aware.