DanielLeens commented on issue #12847:
URL: https://github.com/apache/seatunnel/issues/12847#issuecomment-6249766410

   Thank you for the precise Flink reproduction. Current `dev` confirms the two 
relevant paths: the writer resets `batchCount` after flushing, then its 
close-time merge can be skipped, while the aggregated committer performs the 
CDC merge separately. #12848 is the exact implementation lane and its Build is 
green.
   
   Please keep the fix focused on tracking unmerged CDC data through writer 
close, with the deterministic no-checkpoint E2E reproduction and a regression 
test that proves a final merge occurs only when rows remain. A close-time merge 
failure must remain visible to the engine rather than being silently treated as 
success. Classification: F — Connector-V2 Databend CDC final-merge fix.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to