A watcher node is an Operations Manager agent that runs monitors and rules that test an application or component that is usually installed on another computer. Watcher nodes are used by the following monitoring wizards. You must specify one or more watcher nodes to run the monitor, when you use one of the following templates:
- OLE DB Data Source
- TCP Port
- Web Application
A watcher node can either be the agent with the application or component installed, or it can be a separate agent. If the watcher node is a separate computer, in addition to ensuring that the application or component is healthy, the watcher node can validate that clients can connect to it. The watcher node tests such additional components as security, network availability, and firewalls.
You can best ensure availability of an application or component by specifying watcher nodes in different network segments. For example, a database might be accessed from application servers in each of two different network segments. While the database might be available to one set of servers, an issue such a problem with a network device could make it inaccessible to servers on the other segment. If you use the OLE DB Data Source template to create a monitor for the database with a watcher node in each segment, you are assured that any problem accessing the database will be detected.
You can also specify the computer with the application or component itself as a watcher node. This performs the test without relying on any external components.
For More Information
See Watcher Nodes in the Authoring Guide.