— Start Rant —
I am tired of looking for a simple, functional server monitoring program. I have spent the last week or so trying to get a fully functional setup of Nagios, Cacti, Zabbix, or Icinga and almost all to no prevail. Why is there no EASY and SIMPLE monitoring program out there that is Open Source, customizable, AND functional?
Here’s what I had problems with in each program:
- Nagios: I got it to work, but the web GUI is messy and has way to many un-needed features that get in the way of functionality. Plus, setting up monitoring on external servers takes more work then it should if you ask me.
- Cacti: while nice due to the fact it uses SNMP, for some reason does not graph a single thing properly for me. I guess the max amount of ram in all my devices is always declining… This might be a configuration problem, but out of all of the software choices here this is easily my favorite.
- Zabbix: Don’t even get me started. I spent about 6 hours just trying to add one server to it before I gave up. I can see how it would be nice in a HUGE setup, but the time it takes to add a system is just not worth it to me.
- Icinga: Probably the best for specific service monitoring, but it still falls short. spent the last day or so on it and now that I have it setup for the most part its nice, but it lacks what Cacti offers me… Graphs of load and status over time! Plus, the time it takes to configure and add a server is annoying. They need to make a native reporting server instead of hopping off of the Nagios reporting daemon.
Here is what I request: Someone make an easy to use, all in one Server Monitoring program that can do graphing of information over time like Nagios, but also email you as soon as some service or port goes down like Icinga. Plus, the ability to memo all errors in Icinga really helps with management so that would be nice to see as well. Also, Multi-Platform support is critical.
Looks like I have a new project to work on, because if you can’t find something that works, MAKE something that works. I will probably go with a PHP/MySQL script with a cron file to poll from systems.
*Note: I probably could/should have gave each program a little more time to try and get it working, but I don’t do this for a job, I do it for a hobby, so if it takes me longer then 3 hours just to get SOMETHING working, I will move on.
PS: If you have any recommendations for monitoring software to try out for a small 6 server setup, let me know.
EDIT: Find an update to this post here.