March 2, 2017 at 10:08 am
Just upgraded our disks to 2400 IOPS during the weekend, with logs on one drive and data another. But server is slow again somehow - mostly the ssms takes forever to collapse on object explorer. The CPU is constantly under 50%, total computer Memory usage is at 87%, and I/O reponse is under 10 ms according to Resource Monitor.
Avg Wait in ms:
PAGEIOLATCH_UP 41
PAGEIOLATCH_SH 1
PAGEIOLATCH_EX 34
Our web team complains about slow web responses, too.
What can be wrong? Still the I/O issue? Where should I look?
March 2, 2017 at 10:22 am
You've checked everything, except network traffic and response.
I generally see the same issue when connecting from continent to another, then there's a delay in object explorer visual response.
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
This thing is addressing problems that dont exist. Its solution-ism at its worst. We are dumbing down machines that are inherently superior. - Gilfoyle
March 2, 2017 at 10:39 am
Thanks, Henrico. I did check with our network guy when the system looked bad, but he thought the network was fine. Also, the SSMS was very slow at times even when I RDP'ed into the server. When SSMS object explorer was slow to collapse, sometimes it was still quick to execute sp_who2, but then it can take forever to execute it sometimes, too.
March 2, 2017 at 12:54 pm
After some research, I think I may have tempdb contention problem. Made 8 equal sized tempdb files. Will see how that's going to pan out.
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply