We have been fighting issues with WSUS for almost a year now.
We got so frustrated we spun up a new VM and installed everything from scratch. Still we are consistently seeing 100% of all resources in use for days at a time.
Current specs on the WSUS Server itself: Server 2012 R2, 12gb RAM, quad core 2.4Ghz Xeon
This is fully patched with all the latest updates. So all the previous hotfixes for previous high CPU issues appear to have been included in these hotfixes.
Database is on a separate server running SQL 2016. The SQL server is not showing any issues or stress.
SCCM Site Server (which is on its own VM) is Build 1710. We have ~29,000 endpoints, all on the local network.
IIS WSUS application pool is set with a Queue Length of 25,000, Failure interval 30, Maximum Failures 60, Private Memory 11010048 (just short of the systems maximum, we found it we put to unlimited that we then couldn't do anything to troubleshoot on the server as there were no more resources left).
I ran "exec spGetObsoleteUpdatesToCleanup" on the SQL server and it found none. Its a brand new database so we wouldn't expect to have much clutter at all at this point.
Last time I opened a support case on this MS simply said to throw more resources at it. Our VM Infrastructure Admins are resistant to do so because in their opinion it shouldn't be necessary and Microsoft's own documentation for WSUS indicates it shouldn't be either.
What else can we try? Has anyone else come across this at all? When looking at number of current connections to the server it can vary between 40-4,000 at a time. I have found that even when the connections dropped to ~40 we were still seeing 100% CPU utilization.
Let me know if any additional configuration data is needed. Thanks.
