If you have server visibility enabled on that Windows server it should show you what you're looking for. What you can do is set a custom date range right around that time, and then click on that server from the servers list and then look under the process tab for a list of processes, CPU (%) and Memory should be listed near the end (if not then click on view options and check CPU (%) and Memory, to have them show up and you can filter by those). Remember it averages out based off the time frame you're looking at so make your custom time frame as small as possible.
If the entire thing crashes, you may also be able to find it in the machine agent's logs on that server. In the future you can also setup some alerts looking at specific processes to get alerted if they start to spike too.
... View more
Quick answer as I'm browsing via an intermittent internet connection at the moment.
Are you running this under windows, or a linux environment? For linux, we use the appdynamics service for the machine agent, and have that linked to /etc/rc3.d as a typical service.. I imagine you could do the same via a systemd service as well, though we haven't done so ourselves as of yet. Either way, if you have the service installed, it's pretty straightforward to set up. I'd recommend you start it after your network services though, as it will probably need DNS to function to find the controller, or the agent won't start.
For windows, I believe it's a service so you could set it to auto-start via the services application.
Hope that helps,
... View more