OS Processor Queue Length alerts on Domain Controllers
Hello, we are testing the OpManager product to see if it would be a good fit for us. I've noticed that all of my domain controllers have a high number of OS Processor Queue Length hits. This inspite of the fact that the CPU utilization is less than 5%. I searched the forums here and read through the other posts on this issue.
I decided to run perfmon on my domain controller which had a queue length of 34 and watched it realtime for a few minutes. From what i can see, the OS Processor Queue Length mostly stays low and peaks periodically. Correct me if I'm wrong, but OpManager only polls the system once every 15 minutes (by default). I think because it's reading the current activity when it polls the system, that it is not getting the correct information concerning OS Processor Queue Length.
What needs to be collected is the sustained OS Processor Queue Length so that the average can be determined.
Note the server in question has a single, hyper-threaded processor (2.4 GHz) and 512MB of RAM installed. The CPU utilization is below 10% consistently and often below 5%.
I decided to run perfmon on my domain controller which had a queue length of 34 and watched it realtime for a few minutes. From what i can see, the OS Processor Queue Length mostly stays low and peaks periodically. Correct me if I'm wrong, but OpManager only polls the system once every 15 minutes (by default). I think because it's reading the current activity when it polls the system, that it is not getting the correct information concerning OS Processor Queue Length.
What needs to be collected is the sustained OS Processor Queue Length so that the average can be determined.
Note the server in question has a single, hyper-threaded processor (2.4 GHz) and 512MB of RAM installed. The CPU utilization is below 10% consistently and often below 5%.