What is MRTG, where can I get it?

The Multi Router Traffic Grapher (MRTG) is a tool to monitor the traffic load on network-links. MRTG generates HTML pages containing GIF images which provide a LIVE visual representation of this traffic. Check http://www.ee.ethz.ch/sta/ for an example. MRTG is based on Perl and C and works under UNIX and Windows NT. MRTG is being successfully used on many sites around the net. Check the MRTG-Site-Map.

Round Trip Ping Time to a friend of ours:

Monitoring Scripts (and new stuff)

The monitoring scripts can be run on the local machine and stored on an NFS share, or some other directory. Hourly or whatever is desired, the data can be ftp'd, or pulled by a monitoring server, that runs mrtg and generates the graphs, rather than run locally.

I. Handy script by Mike Glaser who is using a Linux box as the collector, and is monitoring a variety of servers, of which there is an HPUX box. He had to compile tcp wrappers on the HPUX box (uncommented one line in the make file for HPUX) and that worked. Upon editing the inetd.conf file, the Linux box was then able to connect via netcat and pull the HPUX mrtg data. Thanks Mike. Some other tcpd/HPUX links: Stanford Second Stanford Loc. Info on TCPD

II. Added a script that gets data from wget on page load times and size. Can be used to also show latency or load on a webserver, via the load times of the page. Would really just show a correlation with other facts. Also capturing size of the file, can possibly show a hacked size as the index.html page would change size.

III. Added a modified "load script" provided by Schene Groom, it is a better script for capturing low load values, even as low as .02    Schene's script is here.   He is using this on a freeBSD machine. Thanks Schene.
IV. Posting some brief cacti info too:
Example how the df -k script data looks when graphed.


Example of how the process count for specific users looks when graphed: