The IT world has overcome much of its skittishness regarding the use of open source projects. It obviously wasn't always that way, and plenty of IT shops still refuse to consider open source alternatives to commercial solutions. But most folks have come around to the fact that open source tools are a great asset to any organization and won't actually set fire to your data center or ransack the break room, no matter what the sales droids say.
In many cases, the use of open source tools starts in the skunkworks of the IT organization, where a few individuals leverage open source projects to perform a specific task that is either unfunded or underfunded. Once the proverbial camel's nose is in the tent, more open source applications and frameworks find their way into critical IT systems. Yet with little or no advertising, many open source projects that every shop should be testing, if not implementing, never appear on the radar. Here are a few of my favorites.
[ InfoWorld's Savio Rodrigues explains when you should open-source your internal apps | Check out InfoWorld's slideshow of 7 open source projects on the cutting edge. | Keep up with the latest open source trends and news in InfoWorld's Technology: Open Source newsletter. ]
I've been a fan of Nagios for a long time. Nagios is a soup-to-nuts network and system monitoring and notification tool that has an extensive list of plug-ins and a vibrant community. There is a steep learning curve to set it up, but once that's done, you'll have your finger on the pulse of the entire IT plant. You can monitor just about everything on everything: temperatures in the datacenter (through the temperature of each server), latency across WAN circuits, storage capacity, performance, you name it. Implementing a full Nagios monitoring scheme in a reasonably sized organization isn't a one-day project, but it will save significant time and headaches later, guaranteed.
Cacti is a graphing/trending tool that uses SNMP and is a great complement to Nagios. Whereas Nagios can tell you when things go wrong, Cacti can tell you all about the trends relating to that problem. For instance, monitoring storage array capacity with Nagios may tell you that you've eclipsed a threshold, but Cacti will show you the trends related to that array, allowing you to determine the utilization rate over the past six months or a year, which will help determine how much storage may be required in the coming months. This is also true about temperatures, airflow, LAN and WAN links, number of users on a particular server, license utilization (for applications that use licensing tools like FLEXlm), and so forth.
The Web interface is easy to use, and most anything that uses SNMP can be monitored. As with Nagios, there's a large community around Cacti and plenty of plug-ins. The information Cacti provides is only as useful as the length of time it's been running, so if you're not using it yet, there's no time like the present to get started.
RANCID is one of those tools that you'll set up once, make very minor modifications to over the years, and praise for saving your bacon in an emergency. RANCID performs a very simple function: retrieve and organize the configurations of network routers, switches, and firewalls. When set to run every hour, for example, RANCID will email admins when changes are made to any monitored piece of gear, and add the configuration to a version control database that you can then run diffs on to see exactly what changed and when. In the event of a catastrophic failure of a router or switch, you can pull the configuration back quickly and easily, and be sure that it's the latest possible copy.