[mrtg] Re: Problems interpreting decimal "gauge" info (systemload)

[Bernard van de Koppel]

Daniel,

Thanks, multiplying the value gathered via snmp, which effectively makes
it an integer works great. To solve the scaling problem (which
effectively displays the original values) changing the YTicsFactor (for
the scaling in the graph) and the Factor (for the lines below the
graph), deviding the lot by 100 does the trick, displaying the original
values.

Look at the folowing url for the results
http://www.netwerkinformatie.com/mrtg/system-load.html

Thanks again,

Bernard

The altered config used:

# 1 min en 15 min system load
Target[system-load]:
1.3.6.1.4.1.2021.10.1.3.1&1.3.6.1.4.1.2021.10.1.3.3:communityx at 127.0.0.1
* 100
MaxBytes[system-load]: 10000
YTicsFactor[system-load]: 0.01
Factor[system-load]: 0.01
Title[system-load]: System load frans.netwerkinformatie.com
PageTop[system-load]: <H1>Systemload</H1>
Options[system-load]: nopercent, gauge
ShortLegend[system-load]: %
YLegend[system-load]: Load / perc
Legend1[system-load]: 1 minute systemload
Legend2[system-load]: 15 minute systemload
LegendI[system-load]: 1 min
LegendO[system-load]: 15 min



<----------------original message------------>
To: Bernard van de Koppel <bko at netwerkinformatie.com>
Cc: mrtg at list.ee.ethz.ch
Subject: [mrtg] Re: Problems interpreting decimal "gauge" info
(systemload)


I cannot seem to locate it myself in the docs right now, but mrtg
(unless
you are also using rrdtool) needs integer input.

If you need to graph non-integers, you can either:
 a) install rrdtool
 b) multiply the value you get by some number (say 100)
  you would do this by making your config look like this:

 # 1 min en 5 min system load
Target[system-load]:
1.3.6.1.4.1.2021.10.1.3.1&1.3.6.1.4.1.2021.10.1.3.2:communityx at 127.0.0.1
* 100
MaxBytes[system-load]: 10000
Title[system-load]: System load frans.netwerkinformatie.com
PageTop[system-load]: <H1>Systemload</H1>
Options[system-load]: nopercent, gauge
YLegend[system-load]: Load / perc
Legend1[system-load]: Average 1 minute systemload (x 100)
Legend2[system-load]: Average 5 minute systemload (x 100)


Either way will work, although changing the config is definitely the
quicker





Bernard van de Koppel extolled:
>
> Hi,
>
> I have a problem getting the data for my system load, gathered from
the
> local ucd-snmpd (on linux) graphed through mrgt (version 2.9.10 on RH
> Linux 7).
>
> What happens is, that the snmp query will respond with a decimal
vallue,
> but mrtg is not putting it into the statistics log file (only the
first
> line is updated).
>
> Is it possible to poll decimal values through mrtg??? I did not find
> anything about it in the manuals, but found no example also???
>
> Any help appreciated.
>
> Bernard
>
> The first few lines from the "system-load.log" file.
>
> 983372281 1.52 1.75
> 983372281 0 0 0 0
> 983371506 0 0 0 0
> 983371500 0 0 0 0
> 983371200 0 0 0 0
> 983370900 0 0 0 0
> 983370600 0 0 0 0
> 983370300 0 0 0 0
> 983370000 0 0 0 0
> 983369700 0 0 0 0
> 983369400 0 0 0 0
>
>
> The setup I use to poll this value:
>
> # 1 min en 5 min system load
> Target[system-load]:
>
1.3.6.1.4.1.2021.10.1.3.1&1.3.6.1.4.1.2021.10.1.3.2:communityx at 127.0.0.1

>
> MaxBytes[system-load]: 100
> Title[system-load]: System load frans.netwerkinformatie.com
> PageTop[system-load]: <H1>Systemload</H1>
> Options[system-load]: nopercent, gauge
> YLegend[system-load]: Load / perc
> Legend1[system-load]: Average 1 minute systemload
> Legend2[system-load]: Average 5 minute systemload
>

--
--------------------------------
Daniel R. Kilbourne
daniel.kilbourne at voyager.net
Network Engineering
CoreComm - Formerly Voyager.net
________________________________



--
Unsubscribe mailto:mrtg-request at list.ee.ethz.ch?subject=unsubscribe
Archive     http://www.ee.ethz.ch/~slist/mrtg
FAQ         http://faq.mrtg.org    Homepage     http://www.mrtg.org
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi