If running on Windows, monitor the "GeoServer Total Queue Length" performance counter under the ScaleOut StateServer category. This counter tracks how many pending updates need to be replicated over the WAN to your other datacenter. If you see a spike here (say, over 200 for a period of over 60 seconds) then it’s a sign that the WAN is a bottleneck (or the link went down) and can’t keep up with the changes being made in the local store.
