The colo where the hardware is located had announced scheduled 
maintenance for our cluster, but there was not supposed to be any 
downtime ... we are looking into the problem

Dale