Elastic Entity Model
editElastic Entity Model
editThe Elastic Entity Model consists of:
- a data model and related entity indices
- an Entity Discovery Framework, which consists of transforms and Ingest pipelines that read from signal indices and write data to entity indices
- a set of management APIs that empower entity-centric Elastic solution features and workflows
In Elastic Observability, an entity is an object of interest that can be associated with produced telemetry and identified as unique. Note that this definition is intentionally closely aligned to the work of the OpenTelemetry Entities SIG. Examples of entities include (but are not limited to) services, hosts, and containers.
The concept of an entity is important as a means to unify observability signals based on the underlying entity that the signals describe.
Notes
The Elastic Entity Model currently supports the new inventory experience limited to service, host, and container entities.
Enable the Elastic Entity Model
editRequired role
The Admin role or higher is required to enable the Elastic Entity Model. To learn more, refer to Assign user roles and privileges.
You can enable the Elastic Entity Model from the new Inventory. If already enabled, you will not be prompted to enable the Elastic Entity Model.
Disable the Elastic Entity Model
editRequired role
The Admin role or higher is required to enable the Elastic Entity Model. To learn more, refer to Assign user roles and privileges.
From the Dev Console, run the command: DELETE kbn:/internal/entities/managed/enablement
Limitations
edit- Cross-cluster search (CCS) is not supported. EEM cannot leverage data stored on a remote cluster.
-
Services are only detected from documents where
service.name
is detected in index patterns that match eitherlogs-*
orapm-*
.