In a service-based remote support delivery system, service engineers supported by an analysis server receive incident reports from both personal computers and from unmanned servers within an enterprise. The incidents arise both from user-created reports of problems, from event annunciators that monitor hardware and software to report events as they occur, and from the periodic gathering of configuration data. These incident reports are combined with host information and contact information and are transmitted to the analysis server as the central site for processing. All incidents in large enterprises are first collected and stored on an SPOP node. Both proactive and reactive system monitoring is thus combined into a uniform system.