If you have agents deployed to connect to on-premises solutions as we illustrated in Chapter 11, Calling an On-Premises API, then you need to know that the agent running on the on-premises is operating as expected. This is done by each agent keeping a heartbeat signal going every so often. If data is flowing, the ICS server knows the agent is alive and able to communicate. This does mean that there can be a little bit of latency between the agent failing, and it being reported on the dashboard. The following screenshot shows agents both healthy and unhealthy states (as a result of not receiving any communication):
As you can see from the previous screenshot, where agents are grouped, the group is shown with an aggregated status. Unlike the other views, other than using the agent count in the agent group to expand the view to show the child agents, there is not any further drill downs.
Any further diagnostics to determine the issue with the agent will involve a more detailed investigation...
