The cluster configuration I am attempting to construct is not predicated upon throughput. It is predicated upon fault tollerance. Want to setup two independent Redhat Linux Servers that are updated simultaneously by the applications. When one of the Servers goes down the other Server can continue to operate transparently. Once the failed Server is repaired it is returned to the "cluster". The Redhat cluster software implies that all servers must share a common external storage - this not what we want. Have seen some software "Clusterworx" which appears to accommodate this level of independence. Has anyone out there managed to setup two Linux Redhat Servers in such a configuration. Thanks in advance.