May 25, 2016 at 8:45 am
Greetings!
I have implemented CDC in our DB to capture data change for multiple tables and it runs smooth. I know that if I want to include more tables under CDC I can add more child package for populating new tables. But this is good for the incremental load - what happens to the initial load of new table? We know that the initial load should run only once and if I add the new table load DFT in the initial load package then all the other tables included in the package will be repopulated causing data duplication.
Any idea on how to handle this initial loading of new tables added to the CDC implementation avoiding redundant loading of other tables?
May 25, 2016 at 11:16 am
I was able to implement that by adding expressions in the DFT components to make it disable based on project parameter per table.
When I tried the following scenario I found surprise:
I did the initial run with 3 tables and all populated successfully. Then I UPDATED table#1 (didn't run the incremental pkg yet). I then added table#4 DFT to the initial run package and ran it to populate table#4 only. Now, I ran the incremental load package so that the previous update to table#1 shows up in destination table. But NO.... it didn't happen. It seems the initial pkg run to load new table#4 after updating table#1 reset the CDC status and it lost the track for the update.
How to solve that? I should not rerun everything again once I add a new table to the CDC implementation. Any lead will be greatly appreciated.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply