The October 2024 release of Data Ingestion and Replication includes the following changed behaviors.
Optimizing use of source unique indexes when no primary key exists
After the October 2024 upgrade, new application ingestion and replication tasks and database ingestion and replications tasks preferentially use the source primary key when generating Oracle target objects. If a source does not have a primary key but does have unique indexes, the job chooses the best unique index for replication in the following order of priority: 1) the unique index that has only NOT NULL columns, 2) if multiple unique indexes have only NOT NULL columns, the unique index with the least number of columns, or 3) if no source primary key or unique NOT NULL index exists, the unique index with the greatest number of columns.