2 node subcluster, ccm 6.x, cup7.x - users in active active mode. When primary CUP server is failed the CUPC client's that were connected to it manage to connect to the secondary server but we do not get presence availabilty updates i.e on hook/ off ...