2011-11-14

Review: Zenoss 4/5

Have been runnning Zenoss for over a full year, here is my verdict:

Product (4/5):
  • Monitors hundreds of nodes running of a not to beefy vm.
  • Easily extendable.
  • Smooth upgrades. (Not automatic but well documented.)
  • User interface is sometimes a little hard to understand. Not bad for a product the size of Zenoss but not perfect either.
Documentation (4/5):
  • Lots of official documentation freely available covering most of what one needs to know.
  • Forums were migrated to Jive a couple of years ago leaving lots of broken links.
Community (5/5):
  • Nice and helpful community.
  • Lots of contributed plugins a.k.a. ZenPacks. Examples 
    • Hardware health status on HP servers
    • Oracle databases
    • vmware ESX and ESXi
Conclusion:

I did a lot of reading up front before implementing Zenoss and considered 
  • Zabbix,
  • Hyperic,
  • OpenNMS, 
  • Nagios 
 before making the decision. I personally liked both Zabbix and OpenNMS and have heard good things about the but Nagios/Icinga and Hyperic but chose Zenoss because of 
  • agent-less architecture 
  • hp hardware monitoring almost out-of-the-box
  • snmp trap handling.
I recently asked myself  if I'd use Zenoss again and the answer is: Yes, at least if I have to monitor hardware.

However keep in mind that a full implementation might easily take weeks even for a skilled sysadmin, depending on the size of the monitored network, the status of the network (is snmp enabled? Who has to approve it? etc.)