A data ring has two or more disconnectable nodes and a monitoring device for monitoring and driving the nodes. Appliances that communicate with one another can be connected to the nodes. The data ring is distinguished by an additional node that can be connected by the monitoring device. The additional node allows a fault state to be produced deliberately in the other nodes. The deliberate production of a defined fault state in the data ring and the defined propagation of this defined fault state in concatenated data rings results in states being produced in all of the nodes which correctly initiate fault handling routines in the appliances which are connected to the nodes. The fault handling routines are provided as standard and provide the appliances at the node with knowledge of the failure of the faulty node or of the addition of a new appliance at a node.