Keeping up with your network requires consistent care and feeding

Few items in IT are more critical and less appreciated than network and system monitoring frameworks. These (almost) set-and-forget systems deliver tremendous value to any organization; they’re critical to a healthy, stable infrastructure. Yet troubleshooting them may require that you navigate a tangle of legacy configurations and one-off modifications — a crufty endeavor nobody really wants to endure.

That’s why it’s a good idea to overhaul your monitoring and trending systems every few years. It’ll give you a great opportunity to see what’s new. Sure, there’s always the tried-and-true Nagios and Cacti method, but many other players may fit your infrastructure better.

[ Want more networking wisdom? Download the Networking Deep Dive Report by Paul Venezia. | Read Paul Venezia’s post on remote monitoring and control systems, “Stay connected when disaster strikes.” | Get the latest practical info and news with InfoWorld’s Data Center newsletter. ]

For example, I recently decided to give Observium a try. Observium is an open source project aiming to make network and server monitoring and graphing dead easy. The upshot: Once you have an Observium installation running, simply point it at an SNMP-enabled device. It will properly determine what the device is, what data points can be collected and graphed, and what category to place the device.

For instance, pointing it at a Cisco 6509 chassis resulted in bandwidth graphs for every switchport on the device, which you might expect, but it added CPU and RAM utilization graphs for every relevant switching module, all the temperature and power data, and more. Pointing it at a Linux server brought all the expected information down, essentially instantly. There’s no need for manual configurations for supported devices and hosts.

This differs from Cacti and Nagios in a significant way: There are almost no command-line or configuration file modifications required to set up new hosts and monitored devices, nor are there many knobs to turn and buttons to push to customize the hosts. Whereas Cacti provides an extremely thorough and complex method for device addition and data maintenance, Observium shares very little — you tack on a supported host and voila, Observium starts collecting all the information you could ever want on that device.

You can configure just about everything through the Web UI, and it’s designed to be extremely simple. For those who’ve waded through the seas of Nagios configuration file includes, this might be considered a significant advancement.

On the flip side, the list of supported devices and operating systems is necessarily limited. Where Cacti allows for a shocking amount of customization and specific data gathering and graphing functions that can support just about every possible network-connected device, Observium opts to hit the major devices and operating systems — and does them well.

I ran into some bugs in Observium, and the fact that there’s no stable release, just the Subversion nightly builds, seems a bit adventurous for a network monitoring package. Aside from that, it’s worth a try, requiring little more than your normal LAMP stack and maybe an hour to get it set up and configured. Even if it can’t take the place of a large Cacti or Ganglia implementation, it can provide plenty of benefit alongside other tools. For smaller shops, it may be all you really need. It’s open source and easily installed — what’s the worst that happens, anyway?

The other point here is to not allow your network monitoring and trending packages stagnate. Changing, upgrading, and occasionally migrating these tools is a thankless task at best, yet it should be considered critical to the corporate IT welfare. Nobody really enjoys poring over Nagios configurations that have grown unwieldy over the years, but tightening up those systems, reducing clutter, and mothballing elderly information will only help in the long run.

Sometimes we need to take our medicine. If along the way you happen to look around and find a few new projects that can help with the ever-daunting task of network health maintenance, all the better.

This story, “Keeping up with your network requires consistent care and feeding,” was originally published at InfoWorld.com. Read more of Paul Venezia’s The Deep End blog at InfoWorld.com. For the latest business technology news, follow InfoWorld.com on Twitter.