Abstract: A file server system tolerant to hardware and software failures is located over a plurality of hardware nodes. The nodes of the system act as hosts for software components of the system. Several of the software components can be replicated. The replicable software components include the system file store, a checker and a logger. The replicated components have one primary copy and one or more back-up copies. Replica copies of a given replicated component are each located at different nodes. Location and handling of replica copies of a given replicable component is under the control of a replication manager which is a (non-replicable) software component of the system. The replication manager is distributed meaning it can have one of its instances running on each node of the system. These instances inter-communicate to maintain coherence. The failure detector is also distributed, its instances running on each of the nodes, and contributing to an early detection of hardware and software failures.