Details
-
Sub-task
-
Status: Open
-
Major
-
Resolution: Unresolved
-
None
-
None
-
None
-
None
Description
When a CDC table is created, an indexis created on the PHOENIX_ROW_TIMESTMAP(), which makes it possible to run range scans efficiently on the change timestamp. Since indexes always include the PK columns of the data table, additional filtering on the data table PK columns can also be done efficiently. However, a use case may require filtering based on a specific order of columns that includes both data and PK columns, so having support for customizing the PK for the CDC index will be beneficial.