Crossbeam CPU template

[gnaggies]

Crossbeam systems provide detailed CPU performance values via snmp, and since they do so for each CPU core of each module, graphs are potentially numerous and creating graphs manually becomes soon tedious/overwhelming, so I made a data query/data templates/graph templates to simplify the process.

This is my first data query. I have to say I still don't fully understand how the data templates get the OIDs from the data query. My guess was that it is by order, i.e. first data item = first output item from the query, and so on.

I have double and triple checked the accuracy of the graphs by comparing them to manually created graphs and the result was, both are exactly the same, which goes to show that the query-based graphs are reliable.

This is my first contribution to Cacti. Hope this is useful to someone.

Georg Naggies / Vienna, Austria

[gnaggies]

have forgotten to include the actual query, therefore uploaded once more.

also, by the way, exported in Cacti Version 0.8.7e

install by copying the data query "crossbeam_cpus.xml" into the
.../cacti/resource/snmp_queries folder and by importing the other xml files through the GUI.

also, those templates depend on the presence of an "Hourly" RRA

something like this,

Name Steps Rows Timespan**
Hourly (1 Minute Average) 1 500 14400

best described in the tutorial on how to get Cacti to poll at a one-minute interval.