Occasionally, the agent process would shutdown by itself. We have to reach the Application Team and ask the client to let us access the server to start the dynatrace agent service.
Also, everytime we do this we have to go to IIS manager there and add the native module of Dynatrace again. (If we simply go to services.msc and start the service, in agent overview we only see the primary agent but not the secondary module) Any idea why?
Also, second question (this one might be related...or not? No one knows):
INTERMITTENTLY, the host metric (CPU, MEM, NET, DISK) shows no graph/chart although in the ribbon on top of every pane we can still see red and green color bar (and also the reason/explanation for the incident when we hover the mouse to the warning sign).
So it seems like host metric is still being captured, just that somehow...can't render in the chart.
Again, don't know if these two events are related.
Any help on this issue/ these two issues would be appreciated. @Woon Jia J.
Hi Way. Can you tell me which technology we are talking about? I assume you talk about monitoring an ASP.NET Application in IIS - correct?
If that is the case then the .NET Agent will only "shutdown" if IIS is stopping the AppPool Worker Process (w3wp.exe). This typically happens for the following reason
#1: there is an application issue, e.g: memory exhaustion, ... and IIS restarts the w3wp.exe process
#2: there is not traffic for a while which causes IIS to stop the worker process -> that is normal behavior
#3: you have some scheduled recycle events. IIS allows you to specify these recycle events -> check with your IIS Admins
If the ASP.NET Worker Process is not running we might still capture CPU, Memory, ... of your host. Why? Because you might have another agent running on that same machine, e.g: web server agent or a host agent. As long as one agent is running on a host you will receive these metrics
I hope this explains at least some of the things you experience