January 20, 2016 at 1:16 pm
Hi Everybody,
after upgrading our Enterprise 2K8 R2 sp3 instance to new hardware with 1TB RAM and 64 CPU cores we noticed something strange. Our SQL monitor shows that the metric in sys.dm_os_ring_buffers (type RING_BUFFER_SCHEDULER_MONITOR) for CPU SystemHealth/ProcessUtilization is consistently ~3x higher than (100%-SystemHealth/SystemIdle). Basically (if I got this right), it means that for our single instance SQL box SQL process utilizes 3 times more CPU resource that whole system which is nonsense. Overal system CPU utilization seems to be accurate. Clearly the ProcessUtilization is faulty. :crazy:
Does anyone have any idea of what could cause this and what are the consequences?
The instance otherwise looks healthy and definitely operational.
Thank you!
- Anthony K
January 20, 2016 at 1:25 pm
I had the same situation some months ago. I think this article helped me - http://blog.sqlauthority.com/2014/11/07/sql-server-performance-counter-missing-how-to-get-them-back/
Unload and re-load the counters.
Igor Micev,My blog: www.igormicev.com
January 22, 2016 at 10:07 am
Thank you, will confirm as soon as I give it a try next maint window.
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply