A specific thread of the SQL module now delete outdated data from the database to avoid ever growing tables.
Monitoring instances can now be configured to flag all of its services to UNKNOWN when some timeout is reached. This allows you to easily detect instances that have some connection issues and avoid you to visualize unfresh states.
A new statistics system has been integrated in this version. However this is still and experimental feature that has not yet been enabled by default.