Disclosed are hardware and techniques for correcting computer process faults by identifying risk associated with correcting a computer process fault and computer processes that may depend on the corrected computer process. The interdependent computer processes in a network may be determined by evaluating a stream of process break flags from a monitoring component coupled to the network. Each computer process break flag in the stream of computer process break flags indicates a process fault detected by the monitoring component and is correlated to a corrective response. The break flag and the corrective response are assigned a risk. A risk matrix accounts for interdependencies between computer processes and identified corrective actions. A final response strategy that corrects the computer process faults is determined using the assigned risk and computer system interdependence. A runbook stores the final response strategy, which may be updated based on changing computer process interdependencies and assigned risk.