The Reasoning Engine is an event based engine which orchestrates and drives the population of the data model through executing a series of rules that make up the core functionality of BMC Atrium Discovery.

The Reasoning Engine does not consist exclusively of fixed code with associated configuration parameters. The Reasoning Engine can be taught through patterns how to infer data about any environment into which BMC Atrium Discovery is deployed. Patterns drive the instruction of the reasoning engine and the pattern language is designed to follow an Event/Condition/Action ECA architecture just like the rest of the system.

The underlying system is driven by flexible rules that can be customized, changed and added to without requiring a new version of the Reasoning Engine. However, it is not the intention that end users should need to modify these rules.

The Software Instance and Business Application Instance creation processes are now part of Reasoning rather than discrete components and are controlled using the Pattern Language.

Scan level

BMC Atrium Discovery provides the following default scan levels.

• Sweep Scan: This will do a sweep scan, trying to determine what is at each endpoint in the scan range. It will attempt to login to a device to determine the device type.
• Full Discovery: This will retrieve all the default information for hosts, and infer all of the information that we can.
  You can choose any scan level when you are adding a discovery run. If you do not choose a scan level, the appliance default (which is configurable) is used.

Host rescan

The Host page has a Scan Now button. When the Scan Now button is clicked, all IP addresses are shown but the host will be scanned through the IP address determined to be the best IP for that host.

Provenance

When you are viewing a node representing an inferred entity in the UI, you can always access the information which was used to create the entity. Click the Show Provenance button to view the provenance information.

This displays an additional column which contains information for each attribute on how recent the information is, and the discovery method that obtained the information. It also displays an additional row which contains details of the primary data provenance.

For example; on a host, the latest scan From DeviceInfo means that the attribute was updated at the latest scan with information obtained from the getDeviceInfo discovery method.

In the Reports section of the Discovery page, there is a report on Hosts with attribute retrieval failures. This provides information on attributes and their hosts which have were not updated at the most recent scan. See Discovery reports.