Previous incidents

Replicas are now fully restored, our clusters should be working normally now.

This "quick" maintenance update to better our scaling logic for databases turned into a 2 day nightmare, but at last we're done. Thank you for your patience.