[smokeping-users] Graph scale problem

[Bryan K. Hess]

Hi folks,

I noticed a message in the list archives (quoted below) from Wolfgang 
Hennerbichler who seems to be having the same problem I am.

I put a partial screen shot at  http://www.jlab.org/~bhess/scale.png

It seems that for some detail graphs, the smoke portion of the graph has 
different peaks than in other views for the same time and data(!). In 
these graphs, some of the smoke peaks around shortly after 6:00am have 
an 80ms peak in the "Last 3 hours" graph, but a much higher peak in the 
"Last 30 hours" graph.

This is with rrdtool 1.2.27, SmokePing 2.3.5.

I can't explain this, can anyone help?


--Bryan


*Wolfgang Hennerbichler* wh at univie.ac.at 
<mailto:smokeping-users%40lists.oetiker.ch?Subject=%5Bsmokeping-users%5D%20does%20smokeping%20really%20work%20correctly%3F&In-Reply-To=> 

/Tue Mar 25 13:18:49 CET 2008 wrote:/
> Dear Smokeping users, dear Tobi,
>
> I wonder if the graphs I'm getting with smokeping can be correct...
> this is the graphs: http://tiny.wogri.at/JU
>
> You see that the average ping is aroound 400 and 500 useconds. In the  
> last 30 hours view, you also see that the smoke nearly touches the 
> 1ms  border.
> How can it be, that the average smoke of a graph that never is 
> getting  higher than 1ms can be somewhere around 3 and 7 milliseconds 
> in the  last 10 days view?
> I'm sorry if I bother you with standard-question, but to me it just  
> doesn't look correct.
>
> thanks for reading and your time, smokeping-people!
> Wolfgang
>
> -- 
> www.vix.at // www.aco.net // www.sth.ac.at
> wh at univie.ac.at 
> <https://lists.oetiker.ch/cgi-bin/listinfo/smokeping-users> // WH844-RIPE
> Vienna University Computer Center
> Tel: +43 1 4277-14031 // Fax: -9140
>