7 thoughts on “Cacti

  1. I used Cacti on my server before when I hosted my own site. Worked great. Could graph a lot of stuff, not just networking. Is a bit of hard to setup.

  2. I’ll echo the kudos on the Cacti program… and the setup woes. Took me a whole day to make it work properly for all functions, but well worth the time.

  3. We use Cacti at work to monitor our servers, its just a wonderful piece of software. When I first looked into doing it, I was going to just do it manually using RRDTool and writing my own custom scripts to poll the required data from the various servers; then I found Cacti – huge time saver and you can poll anything you want. We’ve got graphing of latency from different servers to different external points, database processes, queries/second and so forth happening. Absolutely thumbs up for a great piece of software.

  4. Just make sure that you are using the latest version. There were a lot of security problems in the previous version

    Beside that it is one of the best monitoring programs, and have a look also at nagios

  5. Setup woes? It’s a click-click web interface. Nagios config = woes. Weeks of tweaking config files kind of woes. But once it’s set up… a thing of beauty. That said, cacti is nice and ‘good enough’ for most. Check out the THold module at http://forums.cacti.net/about6508.html if you want to do some basic threshold alerting in addition to the pretty shiny graphs.

  6. I agree with you Ryan that the configuration of Nagios is more than difficult….

    I use to monitor all the servers and services in my company, and YES it was a pain till i get used to it.

    The THold looks very nice. Thanks for point it out.

  7. I use Cacti on work to monitor my servers. It’s ideal if you have your own machine (e.g. dedicated server or a virtual private server) since you need SNMP to work. Cacti monitors stuff like network traffic, load, processes, disk usage,

    If you have a simple web hosting program, you can still monitor using Nagios, which is very good at monitoring service availability. Nagios can do a TCP check at regular intervals to see if, say, port 80 or 25 is open and responding.