Create a Post
Showing results for 
Search instead for 
Did you mean: 

R80.10 - Integrated SMS,Correlation,SEvent. CPU Overload. 8 Core


have been running and developing an R80.10 SMS for a while now in production, Integrated SmartCorrelation and SmartEvent blades enabled.

On a VMWare platform. The hardrive began to run out of space, so i migrated within VM to a new disk. at the same time, i increased the CPU from 4 > 8 core. and RAM from 16Gb to 32Gb.

however since then i seem to be getting CPU spike issues. specifically, they kick in when i try to run any reports that go over about 7 days.

#top  shows java process hitting upwards of 800% cpu, and 46% of total used RAM.

It will not settle down, cpstop/start resolves. killing the specific process actually kills all log views in smart console.

What specs are others running for this?

or, is there some sort of file i havent modified to account for the new additional resources perhaps?

i was having a gc slave issue. to which i followed the below sk. this has now been fixed (increasing maximum java memory allocation from 2gb to 4gb.). however the cpu spikes still occur. no longer with gc slave, but with a process called ' qtp765773944-51' < multiple spawns of this.

2 Replies

That's not surprising, R80x is more process intensive than R77x versions. SmartEvent also uses significant resources and is usually recommended to run on a dedicated server/VM. The fact that you see spikes when running reports tells me that the system just trying to do what is asked. How many gateways do you have in your environment? How much are you logging per day? What gateway blades are enabled?

0 Kudos

Hi Aaron,

Please see my postings in this thread: 

In particular you may want to have a look at the serversettings.props file, which shows the different breakpoints for Java settings and such based on number of cores and amount of memory detected. 

Updated 2023 IPS/AV/ABOT R81.20 Course now
available at
0 Kudos