Hi All,
Does anyone know the exact flow that the One Agent (full stack) or Code Module (app only) takes to instrument an application.
I am trying to work out the point of instrumentation within the process of loading the agent. What I want to get to, is how essential is the initial communication back to the active gate in the process of instrumentation.
for context, in the logs
We can see the agent load, detects technologies in use, fails initial connection, does some instrumentation, connects on retry, Dynatrace agent config is loaded, agent reports (all within a couple of seconds).
My understanding is that it shouldn't matter about the initial connection failure, since the generally available / default instrumentation libraries are within the one agent / code module images which are loaded successfully and part of the base one agent / code module image. Any additional feature based custom sensors I'd be happy to accept as requiring a connection.
We also have the feature/oneagent-initial retry set.
This is partially validated as it would appear that some instrumentation is missing (not all, not most, just 1 item).
However, the ones I am missing is GA and part of the default java instrumentation.
Hoping someone could help validate this.
I have a support case, but support would have me believe that the entire agent would not instrument anything if the initial connection failed.
