AWS DMS 复制任务

Question

How do I add a new table to an existing replication task and start CDC for it while not impacting/breaking the existing tables on going replication?

Suppose Oracle to Oracle engine. I already have 4 tables in task and cdc is running and in sync. Now I want to add 5th table to it. I did full load for it using native tool(data pump expdp/impdp using flashback_scn) after that I want to start CDC for it. Now do I create a new task for it or add to existing? Can someone please help me understand how to handle such requests.

TIA

Answer 1

You can add tables to an existing task. To add:

1. Stop the task.
2. Modify and add table and source filter (if any).
3. Start the task and choose "Resume". This is critical, if you accidentally choose "restart" instead of "resume", the task will start from the beginning.

What to expect after adding the table: If the task is "Full load and ongoing changes" then the new table will go into FULL Load mode and once done, will move on to CDC.

If the task is "Ongoing changes only" then the new table will start getting CDC data from that point onwards. There will be no full load.

In either case, as long as the task is resumed, the rest of the tables will continue from the last CDC checkpoint and they will not be impacted. I hope it helps. thanks, Naveed.