A method, system, and computer program product to enable other nodes in a cluster to resume operations of a failed node. These operations include storage management services that allow configuration changes to be made dynamically to storage resources. Resource configuration data are synchronized on a set of nodes in a cluster immediately when a resource configuration change is made. If a node that has made a resource configuration change fails, the resource configuration change is available for use by other nodes in the set, each of which can resume operations of the failed node.