Configure fencing in a Red Hat High Availability cluster

When a node becomes unresponsive, the cluster must isolate it to prevent data corruption. Configure fencing since the node cannot be contacted directly. An external fence device cuts off the node’s access to shared resources or performs a hard reboot.

Without a fence device, you do not know if the disconnected node has released its resources. This could prevent services from running on other nodes. Conversely, the system might wrongly assume resources are released, leading to data corruption. As a consequence, data integrity cannot be guaranteed and the cluster configuration will be unsupported.

When fencing is in progress, no other cluster operation is allowed to run. Normal operation of the cluster cannot resume until fencing has completed or the cluster node rejoins the cluster after the cluster node has been rebooted.