Zenoss was founded in 2006, and only two years later, its open source platform has developed a cultlike following in some enterprise network management circles. Zenoss has matured into not only a viable alternative to Nagios, but in some cases an actual replacement for a Big Four system, says Zenoss CEO Bill Karpovich.
Zenoss has managed to grow its customer base from nothing to more than 4,000 installs of its free version, called Zenoss Core. Since Zenoss Enterprise (the subscription-based version) made its debut last year, 132 customers have signed on.
As with any enterprise-class network management system, don't expect to be up and running in minutes. But don't be scared away by the heavy lifting, either: The implementation is simpler than any of the Big Four.
Zenoss offers hardware- and software-based versions of its system. Each platform can run either the Core or Enterprise version of Zenoss, and all versions sit on top of a Red Hat Linux distribution. The hardware-based options are the Zenoss Appliance 250 and Zenoss Appliance 1000, both 1U appliances. The primary differences between the two are processing power and number of nodes that can be monitored.
We tested the Zenoss Enterprise software and found the Ajax-based management console took some getting used to.
The subnet discovery feature in Zenoss allows easy bulk discovery of a large number of Windows servers. We set common group parameters such as the all-important Windows Management Instrumentation user name and password. In two operations, we discovered every Windows box on our subnet and were polling hardware, operating system, service, interface, and performance statistics for all our servers.
Add-ons called ZenPacks add targeted monitoring for specific functions, such as Active Directory, IIS, Exchange, Apache, and Oracle/SQL. The open source nature of Zenoss has led to the community development of a large number of ZenPacks, and as a result, there's a ZenPack to monitor almost every conceivable production system out there.
Zenoss' built-in tiered alerting capabilities can be configured to alert Level 1 staff right away if a service dies. If Level 1 fails to clear the alert within a certain threshold, Level 2 staff can be alerted, and so on.
In our opinion, the coolest feature of Zenoss is its integration with the Google Maps API. There's nothing more slick than seeing your network devices interconnected on a real Google map with full zoom and scroll capabilities. Large environments are no problem: You can create tiered maps so regional IT managers can zoom in on and monitor their own areas.
Zenoss is not for the faint of heart. The interface is less than intuitive and completely Web-based, so plan on investing some time to learn how to navigate around the product. You'll need to rely on community-developed ZenPacks or pay for professional services if you need to monitor a device that Zenoss can't fully model itself. But the vast size of the Zenoss customer base should ensure that device support is added quickly when needed. And according to Zenoss, device support can be added on a support basis for customers in short order if something critical comes up.
The most important two things you'll lose by not buying the Enterprise version are support and WMI collection.
An open source network management system isn't for everyone, but if you're new to the space, definitely give it a try before you take a look at the more expensive alternatives.
|Here Come The Specialists|
|PacketTrap pt360 Pro||Collection of basic troubleshooting and management utilities||$795|
|Spiceworks IT Desktop||Web-based agentless monitoring, inventory and help desk management||Free as adware; ads can be removed for $20 per month|
|Zenoss Enterprise 2.2||Fully functional network management system||Zenoss Core is free; Zenoss Enterprise ranges from $100 to $180 per monitor, based on support|