This paper proposes a distributed multi-agent architecture for wireless sensor networks management, which exploits the dynamic reasoning capabilities of the Situation Calculus in order to emulate the reactive behavior of a human expert to fault situations. The information related to network events is generated by tunable agents installed on the network nodes and is collected by a logical entity for network managing where it is merged with general domain knowledge, with the aim of identifying the root causes of faults, and deciding on reparative actions. The logical inference system has being devised to carry out automated isolation, diagnosis, and, whenever possible, repair of network anomalies, thus enhancing the reliability, performance, and security of the network. To illustrate the advantages and potential benefits deriving from the reasoning capabilities of our management system, we also discuss an application scenario concerning the need of effectively coping with congestion arising in critical parts of the network.