My OMS/OEM repository keeps going out of sync with the agents in the managed nodes. I cannot even ping the agent from the OEM console right after configuration and clean-up of all the q-files, *.jou files, snmp_ro/rw...The nodes that go out of sync and cannot be pinged are on a 4-node RAC cluster. However, all of our single node instances can be pinged and do not go out of sync with the OEM repository. Is there a special configuration for OEM when the managed node is a RAC environment? All our repository 'out of sync' and 'unable to ping agent' problems are only with the RAC nodes!