The disaster recovery framework uses heartbeats to monitor the state between partner clusters. Heartbeats are sent over the public network to detect cluster failures at geographically separated sites. Heartbeat monitoring is part of a partnership configuration. For example, a cluster failure occurs when all nodes of a cluster shut down. Heartbeats could also be lost when a cluster loses access to the public network and no communication occurs between the partner clusters.
The heartbeat monitor uses plug-in modules to query the heartbeat status of its partners. The disaster recovery framework offers default plug-ins for monitoring through a TCP/UDP connection.
The disaster recovery framework enables you to specify a command to execute when a heartbeat-loss notification is issued. You can also specify a list of email addresses to notify when a heartbeat-loss notification has been issued.
You can use customized plug-ins to provide a data path over alternate communication links, such as email, HTTP, satellite, and microwave towers.