1.3.1 Extended Error Data Model

An extended error is one or more error records from an error sequence. Each error record in the error sequence contains up to four values that software agents can use to encode additional information about the error that occurred. These values are called parameters for the error in the error record. For example, if a file cannot be found, which causes a failure in a system, a parameter in the error record might be the name of the file that was not found.

Besides the four parameters, the error record contains the following data elements: a generating component, a detection location, and an error code.

The generating component is a unique numeric value that identifies the component or protocol layer where the error or failure occurred. It is recommended that the generating component be unique within all implementations of this protocol.

The detection location is a numeric value that is unique within a given generating component and identifies the location in the component or protocol layer where the error occurred. Location can be any identifier inside a component or protocol layer that unambiguously describes where the error occurred or was detected. For example, a software agent can assign one detection location for each module or function inside that software agent. Alternately, a software agent can use line numbers to identify the location where the failure occurred or was detected. Any detection location is meaningful only within the context of a specific generating component; thus, the generating component is part of the namespace definition for a detection location.

The error code is an implementation-specific numeric value that specifies the error that occurred.