The Database Maintenance seems to be Incomplete. However, no errors are seen on the Oracle database repository.
The FMS log has errors lines as such,
2016-03-27 23:53:45.580 ERROR [Data-3-thread-14630] com.quest.nitro.service.topology.TopologyService - Failed to create a topology object with name HostProcess: {name=oracle_3xyzxyzxyzxyz}: com.quest.nitro.service.sl.interfaces.topology.TopologyChangeVetoException: The configured limit of 50000 instances of type HostProcess has been reached.
&2016-03-28 03:30:25.529 VERBOSE [QuartzScheduler.PersistencePolicy_Worker-1] com.quest.nitro.service.persistence.management.DatabaseMaintenanceController - The TopologyRollup database maintenance operation finished without completing at 3/28/16 3:30 AM after running for 64.69 minutes. Error message = null
The TopologyRollup runs for too long and doesn't finish processing all the topology objects:
The environment doesn't appear to have a lot of churn right now, but it looks like it had a lot of topology churn/growth in the past.
Increasing the topology limit for HostProcess objects would probably just make this worse since it would increase the topology size.
Follow KB 89950 to recreate the topology tables and remove any churn from the topology history. That should speed up the processing of the topology rollups.
Also, allow the rollup time to run for longer by setting a VM option, e.g,
server.vm.option2 = "-Dfoglight.topology.rollup.min_duration=7200000"
to allow the rollup to run for 2 hours instead of the default 1hr limit.
