Nagios Support Forum

Hi all,

I have a question about graphs when looking at cpu usage in the Graph Explorer for a host.
I cannot exactly explain to one of our customers what these values mean and how they are measured.
In the example there is a clear spike visible in the total_5s graph, but this is not visible in the total_1m and total_5m graphs. (red box).
All values of Last/Avg/Max are pretty much the same, but the graphs shows differently.

I've searched for some explanation how to read these graphs and how they are measured, but couldn't find anything.
Does someone know more how to read these graphs?

Hi,
I searched, but do not find any article regard reading Nagios graph.
From what I understand based on the graph provided, it is just the percent being used / time.

total_5m: CPU percent being used / last 5 minutes

total_1m: CPU percent being used / last 1 minute

total_1s: CPU percent being used / last second

It might be better to place your pointer over each line to get more info.


Regards,
Vinh

Hi Vinh,

Thanks for your response!
The numbers below the graphs however give you an different impression then the graphs showing.
But that was also what I assumed, I couldn't get it confirmed somewhere.

Purely out of interest, do you know perhaps why is chosen for these intervals (1s - 1m -5m)?
The Linux uptime of top commands for instance normally uses the intervals 1m -5m 15m

Kind regards,
Fred

Hi Fred,
Yes, I asked the same question when I saw the graph ... why 1s?

When we talk about CPU, usually we meant work load.

Once second check interval would be way too fast. I'm curious what is their re-try interval before sending out notifications.

I know lot of our customers are using 5, 10 and 15 minutes check intervals.


Regards,
Vinh