Enabling deep monitoring caused a service to become unresponsive with oneagent process cpu spiking quite high. Which makes sense some what, given that we are instrumenting a compiled exe.
In the appmon days we were able to instrument only specific entry points and exclude everything else.
With oneagent I can exlude incoming urls, but I would like to exclude or include specific entry points in deep monitoring.
Perhaps the use of custom services might help you here. Custom services can be created for Java, .NET, and PHP. They allow you to manually define methods, classes, or interfaces as entry points.
The use of process group rules might also be helpful. They allow you to create rules for defining what processes should be automatically deep monitored and detected.
I'll link documentation that further explains each below.
Custom service: https://www.dynatrace.com/support/help/shortlink/custom-services
Process group monitoring: https://www.dynatrace.com/support/help/shortlink/process-group-monitoring
Hope this helps!
There are some valid use cases once in a while which required that we excluding certain classes from instrumentation because they dont work well with the OneAgent. Would like to have this feature.