AWS DMS - intended way to work with replication tasks

Question

I'm using AWS DMS to CDC changes on MySQL database and load data into an S3 bucket. On MySQL database, there are a great number of tables, some being huge both vertically and horizontally. As I do not need all the tables with all the columns, I create a replication task with transformation specifying the table(selection) and set of transformations(include-column) specifying only columns that I need at the destination.

I do not understand, what happens when I need a new table from the source. Let's say I update the replication task adding that new table, and now I need old tables to continue CDC, and the new one I need Full Load First with CDC following. I'm not sure what is intended way to handle such a situations?

Answer 1

If the new table matches the table mapping rules you created when you defined the task then you don't need to do anything. New tables will automatically be added to DMS and sent to the target endpoint.