Script was updated due to initialization errors. Please replace old one
Templates have been updated due to errors with newer cacti versions
This post is based on
http://forums.cacti.net/viewtopic.php?t=4028 which contains the well know apachestats XML of Rolf Poser. Due to several questions arising (and some fun to have yet another simple perl script), I extended both two of the existing Graph Templates as well as implementing two new ones.
I did not include the docs from Rolf, as they hold for this version as well.
The script is based on the ws_apachestats_hs.pl, so it uses wget. This may as well be adjusted at your own. It has been enhanced in two ways:
- it includes CPULoad and Thread Scoreboard output
- output is performed in one single "print" statement to overcome possible problems using cactid
With later cacti versions, problems were reported with two of the graphs, the "Bytes/Requets" and the "Thread Details (%)" one. These have been corrected as they were due to the fact, that LINEx statements without color were defined.
Explanation of the Output for the "Thread Scoreboard" is best done using the "server-status" URL at your Webserver (e.g.
http://<your webserber>/server-status)
ATTENTION
This script extends the Apache Webserver Data Template. So be careful, if you are using the apachestats-0.4 already! As well, the existing Apachestats Graph Templates will be modified!
At the moment, I do not know any way, to modify existing RRDs to include the additional Data Sources needed for CPULoad and Thread Scoreboard (perhaps export/manually modify/import will do?)
SO IT IS WISE NOT TO IMPORT DIRECTLY TO A PRODUCTION SYSTEM
If anyone knows a better way to do this, I'll appreciate to here from you.
Multiple RRDs
Due to the lack of my knowledge of the cacti system, I was not able to suppress the generation of one RRD for each of the Graph Templates. So, if you are going to define all 7 Graph Templates, you will notice 7 RRDs as well. All of them will hold "the same" data. The webserver is inquired 7 times for the same set of data. This is not a desired behaviour. There is a chance to get rid of 6 of them manually (Well, I did this yesterday, but took me "some time". It seems not that easy to describe this here)
As always, use this at your own risk.
Comments, enhancements and the like are appreciated
happy cactiing
Reinhard