Services 'overall' graph should show multiple traces (one per sitemonitor)

Posted on behalf of a customer:

The services \'overall\' graph should show one status trace per in-use sitemonitor location, in much the same way a multi-instance datasource \'overview\' graph does. Currently the \'overall\' graph just shows a 0/1 trace, in order to find what the problem is and which site monitors are reporting it, you have to read all the alert notifications or expand each location graph in turn.

A secondary, related, request is for the \'overall\' graph to display when any problems occur (rather than only when an alert is declared based on all / more-than-half, etc, locations reporting an issue). If the \'overall\' graph worked as a datasource \'overview\' graph, this would address both requests with one change.