Zenoss Core

Zenoss Core’s faultless and quick discovery of all our devices, servers, clients, operating systems, applications and running processes was a delight to behold.

On an ongoing basis and on a schedule we set, its discovery process automatically detected and logged changes to our network, including server configuration changes – a feature that even most commercial products lack.

The related discovery-change reports showed helpful historical node, subnet or entire network-level inventory detail. Moreover, a Zenoss Core threshold monitors these changes and can issue an alert when the network configuration changes. However, Zenoss Core doesn’t support as many diverse devices as HP OpenView or Argent Extended Technologies, nor does it monitor Microsoft Exchange or SQL Server as closely as a commercial tool does.

Even more impressive than its discovery of our network is its remediation features, which can automatically execute start or stop operations for a Windows service, for example. Like most commercial monitoring and management products, it also ran custom scripts we wrote that, including one that deleted files when available disk space fell below a threshold we set.

Zenoss Core monitored almost every aspect of the network. It used ICMP pings and SNMP to track device status and responsiveness. It examined network traffic flows to monitor standard TCP/IP services, such as HTTP, SMTP and FTP.

It kept an eye out for port use. It knew which background services and processes were running in each Windows server as well as which processes were running in our Linux machines.

For network outages affecting multiple downstream subnets, Zenoss Core behaved just like a commercial monitoring tool to rather smartly use Layer 3 network-dependency data that it had collected to pinpoint the router or other device whose failure caused the outage.

For any of these monitored items, we set thresholds that Zenoss Core used to alert us of problems and potential failures. The product even has an internal monitoring capability to alert you if Zenoss Core crashes or misbehaves.

Zenoss Core alerted us to threshold breaches via e-mail and pager. If we ignored these alerts, Zenoss Core escalated the problem by e-mailing and paging a second tier of administrative support people we had set up. Zenoss Core’s escalation facility and rules are advanced enough to rival those of monitoring and management tools from such vendors as CA, IBM and HP.

Zenoss Core reports show global inventory, network change history, device detail, network problems and network availability. For time periods we specified, it displayed graphs of network and server performance data that we could use for capacity planning.

We appreciated the ability to specify service-level agreement (SLA) parameters in Zenoss Core and then have the product track SLA compliance for a particular group of devices, servers and applications. The user interface, which sports a customizable dashboard that categorizes network events by severity and by business function, is easy to navigate.