Abstract: A distributed network management system for managing redundant global network control points. The network management system allows candidate control points to determine which is primary and which are alternates, initializes the primary and alternate control points, synchronizes global data, and manages client interaction with the control points. The control points negotiate primary status by establishing a named resource in each processor and by testing for the lowest network address in each active processor. In addition, each client workstation determines which of the candidate control points is primary by interrogation. Client processes detect primary control point failures and switch the alternate control point to primary status.

Abstract: A system for providing distributed control of a resource with centralized configuration management. A network of distributed workstations is provided for controlling a resource such as a large mainframe computer. Server workstations for applying control commands to the resource are attached to each resource, preferably through redundant connections. Remote client consoles are defined which may be connected to the server for control of that resource. Server and resource location and primary and fallback connection paths are maintained by a centralized control server. Upon client console request, the central control server causes the server workstation associated with a particular resource to establish a control session between the server and the client console. A record of each dynamic connection is maintained by the central control server. Upon failure of a client console, the server accesses central server fallback information and establishes a fallback connection with an operational client console.