Looking at the thread dump it does not appear that the FMS failed to start, rather it is taking a significant amount of time attempting to load topology objects from the database. After restarting the server, FMS was able to startup successfully and user is able to login to Foglight console.
1. Check for Topology churn, by collecting the sizes of all tables in the Foglight DB that contain “topology” in their name. If any of the tables are too large e.g. more than 2 GB, then you may need to perform topology table cleanup procedure outlined in KB 89950.
2. Check for DB performance bottlenecks if this is an external DB.
