Identification activities to match instances across datasets

In a reconciliation job, you create an Identify activity to establish a common identity for instances in two or more data sets, confirming that they are the same real-life configuration items (CIs). The identity is stored as the ReconciliationIdentity attribute in each instance. 

Only instances that have reconciliation identities and for which the ReconciliationMergeStatus attribute is set to Ready to Merge can participate in a merge. After an instance is merged into the target data set, the ReconciliationMergeStatus attribute is set to Merge Done.

The identity of an instance in the production data set for that identify activity is copied to matching instances in the other data sets. The production data set is typically the same data set that is the target of your merge activities. If you plan to use independent merge activities to merge each of your source data sets into the target, also use independent Identify activities, each of which identifies one data set against that target. 

An Identify activity assigns an identity to an instance according to one of the following scenarios:

• If one or more instances from the staging data set such as BMC.ADDM or BMC.SAMPLE, match an instance in the production data (BMC.ASSET) set that already has an identity, that identity from the production data set instance is applied to the other matching instances in the staging data set.
• If one instance matches an instance in the production data set that does not already have an identity, the reconciliation engine automatically generates an identity and applies it to the matching instances in the other data sets. The reconciliation engine can optionally also apply the identity to the instance in the production data set.
• If an instance does not match any instance in the production data set and its own data set is marked for automatic identification, the reconciliation engine automatically generates an identity for the instance.
• If an instance does not match any instance in the production data set and its own data set is not marked for automatic identification, the instance remains unidentified.

For information about creating a standard or custom reconciliation job with an identify activity, see Standard reconciliation jobs and Recommendations for customized reconciliation job.