So I started to play with SNMP a while back. I found I could use it in conjuction with MRTG to make nifty graphs of system information like network and disk usage and anything else you can break down to a number that you may want to chart.
I worked at getting a disk usage chart together from SNMP but it just didn't want to work how I wanted it to. Some of the problem seems to have been SNMPs way of handling/displaying large files systems or not displaying them properly as an integer because the sizes were too large for that kind of thing. I kept searching about the internet for more ways to do what I wanted. I found a perl script and a couple other round about ways that I had trouble modifying or was too lazy to do.
So I wrote up my own shellish sript to do it all in MRTG.
This is how i did it. I realize i coulda removed a step but it works.
Target[imageshome]:`df /mnt/images | grep % | grep -v Use | awk '{print $5}' | awk -F% '{print $1}' ; df /home | grep % | grep -v Use | awk '{print $5}' | awk -F% '{print $1}'`
MaxBytes[imageshome]:100
YLegend[imageshome]: % used
ShortLegend[imageshome]: %
Unscaled[imageshome]: ymwd
Legend1[imageshome]: images
Legend2[imageshome]: home
LegendI[imageshome]: images
LegendO[imageshome]: home
Options[imageshome]: growright,gauge,nopercent
Title[imageshome]: Server % Disk space used on home and images
PageTop[imageshome]: