You might be getting stuck by the oddity that when you specify a custom
create command it changes the RRDGRAPH from the deprecated format to the
new format. When Zenoss does its own graphs (at least in Zenoss 1.1.2),
it uses the following format for designating the vname (ds0):

DEF:ds0=/opt/zenoss/perf /Devices/webserver/checkWebsite1_time.rrd:ds0:AVERAGE
LINE:ds0#00cc00:time|
GPRINT:ds0:LAST:cur:%5.2lf%s|
GPRINT:ds0:AVERAGE:avg:%5.2lf%s|
GPRINT:ds0:MAX:max:%5.2lf%s\j

it then uses the CF definition in the GPRINT statement to grab the value
on the fly. The current way of specifying this is to use the following
format, where the vnames are designated by VDEF statements

DEF:checkWebsite1_time=/opt/zenoss/perf 
/Devices/webserver/checkWebsite1_time.rrd:ds0:AVERAGE
LINE:ds0#00cc00:time|
VDEF:checkWebsite1_time_last=ds0,LAST|
GPRINT:checkWebsite1_time_last:cur:%5.2lf%s|
GPRINT:checkWebsite1_time:avg:%5.2lf%s|
VDEF:checkWebsite1_time_max=ds0,MAXIMUM|
GPRINT:checkWebsite1_time_max:%5.2lf%s\j

You will need to look to see what VDEF statements are being used in the
Z2.log file when displaying your custom graph. When you start using
custom graph statements, you should see a statement along the lines of:

|VDEF:datasource_datapoint_maximum=datasource_datapoint,MAXIMUM|

where:
datasource_datapoint_maximum is the name you selected in data source
selection field of the graph configuration, with _maximum appended. This
is what is being used to display the actual max: field in the legend. 

You should be able to graph the maximum by adding a LINE statement to
your custom graph stating:

LINE:datasource_datapoint_maximum#color:mymaximum

where mymaximum is the label name.

One word of caution. I'm not sure what data source you are graphing the
bandwidth of now (last I saw you were trying to get multiple check_http
time sources to graph). Be aware that MAXIMUM in this case, is the
maximum value of the data source recorded. It would not be the maximum
bandwidth available. I may have missed a thread somewhere, but what data
sources are you using to use to construct your graph, and what formula
are you trying to use to calculate bandwidth usage? I don't think core
hours should have any affect on your calculation, as long as the
performance monitor is collecting data continuously. 


 

On Fri, 2007-06-29 at 09:37 +0000, Sator81 wrote:
> Very nice jamesroman  :) 
> 
> >From your guide i tracked the rendering down to the file 
> >/opt/zenoss/Products/ZenRRD/RenderServer.py, i think.
> 
> But when I change the CF type from AVERAGE to MAX no change ocures. But there 
> is also a .pyc version of the file wich from the looks of it seems like a 
> compiled version off the file. Does that mean Zenoss only reads the compiled 
> version so I will somehow have to recompile RenderServer.py?
> 
> The reason for my need of this is that we report to customers bandwith usage 
> per month. And when the data gets averaged out over a month it only displays 
> like 1/3 of the actual usage since core hours is just between 07:00 and 17:00 
> on weekdays.
> 
> ------------------------
>  Kai
> 
> 
> 
> 
> -------------------- m2f --------------------
> 
> Read this topic online here:
> http://community.zenoss.com/forums/viewtopic.php?p=8201#8201
> 
> -------------------- m2f --------------------
> 
> 
> 
> _______________________________________________
> zenoss-users mailing list
> [email protected]
> http://lists.zenoss.org/mailman/listinfo/zenoss-users
-- 
James D. Roman
IT Network Administration

Terranet Inc.On contract to:
Science Systems and Applications, Inc.

_______________________________________________
zenoss-users mailing list
[email protected]
http://lists.zenoss.org/mailman/listinfo/zenoss-users

Reply via email to