We have observed multiple cases of the application of the stock system monitors on Windows 2008 R2 endpoints resulting in unacceptably poor performance on the endpoint. I was unable to identify any specific processes or services as the culprit, but non-service CPU utilization was persistently at 80+%.
In the absence of an substantial response from Kaseya support, I was able to resolve the issue by manually stopping and removing the Kaseya-created performance monitor objects and rebooting (and by removing all Kaseya monitors from the affected endpoints). On one hand, this leaves us unable to use Kaseya to monitor performance on the affected endpoint. On the other hand, the ONLY performance issues we've experienced on this endpoint were caused by Kaseya monitoring.
Has anyone experienced anything similar or does anyone have any thoughts?
I have occasionally experienced this, and like you, was unable to trace it to anything substantial except in one case where it was due to interaction with the client's antivirus product. In all cases except that one, it was solved by a removal and reinstallation of the agent on the endpoint. Because that was an easy fix, and I can imagine it being relatively nightmarish to debug, I've never raised a ticket for it. If you have a server you can produce the fault on and tolerate leaving it in that state while it's investigated by Kaseya Support, it would probably be beneficial to raise a ticket.
I have seen it on every 2003 server I monitor. I it leave handles open. I had to write an agent procedure to restart the Performance and Alerts Services daily.
We also have been having serious issues with 2003 servers (seems to be just physical servers if that means anything). Great idea about restarting those services.
Linus, have you opened a ticket that I can reference when I open mine? Anyone else open related tickets?
FYI - When I have seen this, it's usually because I have an old agent installed. Go to the agent tab and update the agent to the latest and greatest and see if the issue is still occurring (force update it please).