Developer's Guide
The Zenoss system brings together many types of monitoring and management information. The information is available through a standard web browser. In fact all aspects of the system are accessed though the web there is no need to edit configuration files.
At a high level, Zenoss consists of four major parts:
Model
Availability
Events
Performance
At the core of Zenoss is the Model. The standard model is a detailed description of all the devices Zenoss manages and their relationship to your business or other important groupings. Because of the large amount of information in the model there are several ways that information can be added. First is through auto discovery. This is the primary way that information is added to the model. Zenoss auto-discovery is very flexible and can use several different transports. The model can also be populated though the web UI or through Zenoss' external APIs. Version 2.0 adds discovery locking which allows auto-discovered information to be overridden with manually added information.
The model is used to drive the monitoring elements of the Zenoss system which will be described throughout the rest of this document.
Availability monitoring consists of running tests against the IT infrastructure to determine if it is currently functioning properly. These test are typically run externally to the monitored system. Example tests include: ping, process, and service tests.
The Zenoss Event Management System is a consolidation of status information from all parts of the Zenoss system as well as external systems. When a Zenoss monitoring daemon detects a failure or threshold breach events are generated. This is similar to most other monitoring systems available. Zenoss does more in that it also takes event import from other parts of the IT infrastructure. These include Syslog and SNMP Traps. It's one thing to bring the events into a single repository but an event management system must do more. As events are received Zenoss runs them through a set of rules that augment the information they contain and integrate them with the model.
The Zenoss Performance Management System tracks important IT resource information as it changes over time. This process is also known as data collection. It is critical to know how much disk space is available, what the CPU load is and how long a web page takes to download. This system can collect information though SNMP, custom scripts (ZenCommands) or XML-RPC. Performance information is integrated with the Zenoss Model so that resource usage is shown in the context of other Zenoss information.