Hi,
In that scenario 2 in the documentation, there’s a Cluster ActiveGate deployed that’s listening to tcp/9999 for synthetic monitoring data incoming from various public locations hosted by Dynatrace. I’m not sure whether “Frankfurt” in your example is hosted by you or Dynatrace? Because both options are actually possible.
So to clarify: you can execute synthetic tests from various servers around the world which are hosted by Dynatrace. If your target app is available from the internet, it’s likely a good idea from costs perspective to use the existing monitoring locations that Dynatrace provides. If your application is in a private network, you probably want to monitor from within that same network and host your own monitoring server. So these are the options:
Option A: When using one of the public locations hosted by Dynatrace, you need the Cluster AG that has tcp/9999 open for incoming connections from the internet (so it can receive data from the various synthetic monitoring sites hosted by Dynatrace). Note that the direction is not outgoing from the ActiveGate, it’s from the Dynatrace monitoring locations towards your ActiveGate tcp/9999.
Option B: When using a private location hosted by yourself, you just need to be able to send the monitoring results towards your Managed cluster port tcp/443. (or, if needed, you can route the data via proxy or a Cluster ActiveGate).