Video Screencast Help

VCS 6.0.1 linux - How CPU usage is calculated?

Created: 07 Jun 2013 | 6 comments

Hello All,

Maybe this is already known answer but I tried to find how VCS 6.0.1 HostMonitor is calculating CPU usage on Linux hosts and didn't found answer.
Can somebody of you answer this question?
I would like to know from where it is collecting CPU data, if it is an avarage CPU usage... (It is looks like that collected information is wrong)

Thank you very much.

Iv4n

 

Operating Systems:

Comments 6 CommentsJump to latest comment

mikebounds's picture

It is an average CPU usage - see https://www-secure.symantec.com/connect/forums/cpu-usage-exceeds#comment-8616191

Mike

UK Symantec Consultant in VCS, GCO, SF, VVR, VxAT on Solaris, AIX, HP-ux, Linux & Windows

If this post has answered your question then please click on "Mark as solution" link below

Iv4n's picture

Thanks for your answer Mike.
But do you know how it is collected on Linux? (I saw this thread before - but it is for Solaris - So I am expecting the similar answer for Linux)

Actaully We have problem that from engine_A.log we can see that CPU usage was on 100% but another monitoring show us that it is not truth. So I would like to know from where it is collected and probably some details about HostMonitor.

Iv4n
 

g_lee's picture

According to this technote:

http://www.symantec.com/business/support/index?pag...

Linux
Compare the usage reported by HostMonitor with the info found in the /proc/stat file

If this post has helped you, please vote or mark as solution

Iv4n's picture

Thanks :) Well I saw also this technote. But it is not answer which I am finding.

I know that I can compare it with another tools (and also I have done it) but I would like to know how it is collecting data.
(what is using HostMonitor for data collection)

I have checked all threads found on google :)

Faisal Saleem's picture

Dear lv4n

Hope you get the concept behind, to find your answer.

Basically the CPU usage reported by HostMonitor agent is an average value over the MonitorInterval period of the usage of all CPUs. So the value is a average value during the last period of MonitorInterval secs, and combined all CPUs usage.

Hope this would be helpfull.

 

Setu Gupta's picture

On Linux, HostMonitor uses the "/proc/stat" file to determine the aggregate CPU usage. HostMonitor reads the first line to get the time spent by all the CPUs in the system after every MonitorInterval seconds (which is 30 by default). It reports the CPU usage based on how many jiffies were spent idling and how many were spent in performing some tasks since the last time /proc/stat file was read.

HTH.