Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I've been using Nagios for a while now and recently started using Cacti after being dissatisfied with the lack of scaling and lack of any GUI in MRTG. I'm interested in adding trending to my Nagios installation and wondered what was the best route to go. I've looked around a bit and have seen what's available, but there's not a lot of information around to differentiate them from each other. My Nagios install has about 250 hosts and 1100 service checks, but many of them are just simple network devices and there's only about 20 servers and 300 services associated with them. All servers but 2 are running Windows Server 2003. What are the main highlights of PNP4Nagios vs. nagiosgraph, or would I be better off using some sort of tool to convert the data to RRD form and just view it directly in Cacti? Is there a completely different direction I could go that would be even better? Please comment if you need any more information, I tend to be too wordy and tried to keep this question brief.


share|improve this question
up vote 6 down vote accepted

I ended up going with pnp4nagios. Installation was a breeze and there was no additional configuration needed to get it automatically graphing trends for every service (that produces graphable data) on every host. The entire process took maybe 30 minutes.

share|improve this answer
Definitely, I recommend grabbing the 0.6x version. BE AWARE The installation process and therefore the documentation is different. Found 0.6x to be much easier to configure than 0.4 – Dan Mar 12 '10 at 16:31
I did end up going with the 0.6x version. I followed the documentation for that and found the whole process to be very straightforward. – Matt Mar 22 '10 at 14:41

did you try installing nagiosgraph? any feedback you have about the process would be appreciated.

nagiosgraph has seen a lot of updates in the past couple of months, but there are always ways to improve. some things added in nagiosgraph 1.4.x include:

  • more ways to slice the data
    • all services per host
    • all hosts per service
    • selection of individual data sources
    • aggregation of multiple data sources
  • better ways to view/navigate the data
    • embedding in nagios
    • popup graphs on mouseovers in nagios
    • appropriate links between graph pages
    • configurable time periods and end dates
  • authorization for specific hosts/services
  • out-of-the-box experience
    • ensure default rules match 'standard' nagios plug configurations
    • simplify installation as much as possible
  • documentation
    • more examples
    • detailed installation and configuration document
share|improve this answer
I did not. Pnp4nagios seemed to be the general consensus of the internet at large so I tried that first. Since that went so smoothly I wasn't really interested in trying a bunch of alternatives. – Matt Mar 22 '10 at 14:42

I highly recommend Munin if you want a standalone tool that's easy to set up and extend; you get a good amount of trend graphs enabled out of the box and there are plugins for pretty much anything else you might need. Munin is RRD based but stands alone from Nagios; installation is pretty much analagous to Nagios, you install a central server and a daemon on each node you want to monitor (via package install on most Linux distros) and then optionally do a small amount of configuration on the nodes if you want to add non-default data.

share|improve this answer
I forgot to mention in my initial post that most servers are windows, not *nix. I edited the post to add that. From a quick glance at Munin and the FAQs, it doesn't really seem to support windows clients. Am I off base here? – Matt Feb 23 '10 at 21:10
There is a Munin node daemon for Windows, or you can use SNMP - see I've only used Munin on Linux, so my "easy to set up and extend" comment may or may not apply ;-) – gareth_bowles Feb 23 '10 at 21:55

I tried Zenoss and to start with it looks quite impressive, but unfortunately that's where it ends. Underneath it all you're left with a system that isn't very useful, the reports aren't useful at all and when you start adding more than about 50 devices it starts slowing down to a point where it's almost un-usable.

share|improve this answer

*disclaimer, I'm the Zenoss Community Manager

Zenoss does exactly what you are looking for. It is an Open Source network monitoring platform that does availability monitoring (like Nagios) and performance monitoring (like Cacti). It's built to easily scale to thousands of devices and does Windows machines via WMI and SNMP. There are no configuration files and you get your monitoring, graphing and it's easy to extend. You can also re-use any custom Nagios and Cacti plugins you have and there are no agents to install on the nodes.

share|improve this answer
I'm not really interested in replacing nagios, and not interested in something that either is or is related to a commercial product, but thanks for mentioning it. – Matt Feb 24 '10 at 18:52
@Matt - nagios is related to commercial products, too: as are all other "full-size" monitoring tools I know of (btw, OpenNMS is identical for the community and supported versions, unlike Zenoss) – warren Mar 10 '10 at 19:20
@warren - What I meant was I'm not interested in a crippled version of something whose main product is commercial, such as Zenoss. How is nagios related to commercial products other than the fact that some products use it as their engine? – Matt Mar 22 '10 at 14:49

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.