Windows System consuming 80% + of CPU Resource, how to further troubleshoot in Dynatrace?

[waikeat_chan]

Windows System consuming 80% + of CPU Resource, how to further troubleshoot in Dynatrace?

(other than asked customer to open task manager to have a look)

Best,

Wai Keat

[ChadTurner]

That's a good question. I don't think you can break 'Windows System' out further in Dynatrace. 

[kalle_lahtinen]

Yep, which makes alerting about it a bit problematic. We also often see "Packet retransmission rate for process Windows System on host asdf has increased to 10 %". Seems like the best option is to just suppress these alerts, because there's nothing you can do with that information. ...Or if someone has ideas, do let me know 🙂

[AntonioSousa]

One idea might be to look at the Windows logs at the host level in Dynatrace. Might find an indication of something that might be going on. Also, take a look at the number of DNS queries, as it has surprised me in some occasions.

[Yosi_Neuman · Accepted solution]

Hi @waikeat_chan 

There is a way to figure out which processes are include under windows system from SA.

Go to diagnostic files --> Agent --> logs --> os and open processesSnapshot.log 

In this file look for Group name: Windows System, under it you will find all the windows processes that OA collect together.

You will see for each process last CPU val and Virtual memory peak size and can get a clue on which process is consuming more CPU and Memory than others. 

HTH

Yos 

[lawrence_cuneaz]

To get a process snapshot you can enable automatic process snapshots that run based on any of these triggering events or you can run a snapshot manually by selecting ... in the Process Snapshots section on a specific host page.

• High host CPU usage
• High host memory usage
• High packet drop rates
• High NIC utilization rates
• High number of NIC errors
• Process availability events