Arul, I have some experience of synchronous CDC in 9i. Not sure whether its exactly the same in 10g but here are some comments: - > 1. After letting dbms_cdc_publish.create_change_table create the change > table, is it ok to (manually) add columns to it? I need to add a couple > of > snapshot columns that are not present in the base table. I think that would be a bad idea. There is an ALTER_CHANGE_TABLE procedure in the DBMS_LOGMNR_CDC_PUBLISH package for this purpose. > 3. Does Async CDC partition the change table automatically? If so, what > is > the partitioning scheme? (The manual simply tells "Change Data Capture > automatically manages the change table partitioning as part of its change > table management") It creates the change table with one partition. I think that it messes around with the partitions when running the PURGE procedure. Regards David Lord -- David Lord davidclord@xxxxxxxxxxxxxx -- http://www.fastmail.fm - The professional email service -- //www.freelists.org/webpage/oracle-l