Hi,

We have approved Augusts updates on monday. Since then all targeted systems CPU is peaking to 100% for 15-20 minutes on every "check for updates" interval... in our case every 2 hours. svchost (wuauserv service) and TrustedInstaller are the culprits.

I can already confirm the issue occurs on virtual machines (vmware), running a perfmon trace on a physical system now.

Anyone experiencing this issue also? Or any idea what might be causing this? I know it is normal behavior Windows checks for updates and the cpu peaks briefly but a peak of 20 minutes every interval doesn't seem normal to me.

Grts. 

• Edited by BallieWallie Thursday, August 20, 2015 2:31 PM

We are also experiencing the same issue as are others as noted in other threads in this forum.

https://social.technet.microsoft.com/Forums/windowsserver/en-US/8458ab49-9816-4717-8d72-8949477186af/check-for-updates-takes-30-minutes-on-windows-2008-r2-sp1-servers?forum=winserverwsus

https://social.technet.microsoft.com/Forums/windowsserver/en-US/77d7396f-6032-4f62-90ec-85bc55ac90f8/high-cpu-while-running-wuauctl-detectnow?forum=winserverwsus

We already have a case open with Microsoft.  I would recommend raising a case and in that explain others in the industry are having the same problem so it's not assumed it is limited to your environment.

If you have any feedback or solution please do post back.  I will do the same.

Isnt this a intended behavior?!

Looks like the client is updating his update list.

You could tweak your WSUS a little bit and removing superseeded updates etc.

Link

I experienced this behavior lots of times in the last company. 

I have set the "Automatic Updates detection frequency" to 22 hours thus limiting this behavior to once a day.

It is indeed intended behavior Windows updates his update list but the peaks in cpu usage we are having since the last rollout is not normal. We haven't had this issue before (see chart). 

Grts.

After working with Microsoft support and further clearing out superseded updates (declining them) we were able to resolve our issue.  Average search time went down from ~40 minutes to less than 1 minute.  Our problem was related to WSUS and superseded updates.  We had a large number which needed to be declined.  Therefore it may or may not relate to the case here even though we experienced high CPU on our clients.

The following script helped us to resolve our issue.   Make sure you back-up your WSUS DB first.

http://blogs.technet.com/b/configurationmgr/archive/2015/04/15/support-tip-configmgr-2012-update-scan-fails-and-causes-incorrect-compliance-status.aspx

The reason we ran into this in August was because one of the released updates MS15-080 superseded a very large number of updates.

• Marked as answer by Steven_Lee0510Microsoft contingent staff, Moderator 1 hour 22 minutes ago

Hi,

Thanks for the sharing!

Your time and efforts are highly appreciated!

Best Regards.