* Fix race in TimestampDataUpdateCoordinator
The last_update_success_time value was being set after the listeners
were fired which could lead to a loop because the listener may
re-trigger an update because it thinks the data is stale
* coverage
* docstring