How to Use Online Refresh for Initial Load or Full Re-initialization
This procedure assumes that the replication jobs are fully initialized and capture time has been positioned before the start of the oldest open long-running transaction on the source. The full load can be done for all tables in the channel or just for selected tables.
Instructions
This is the recommended procedure:
Open the Refresh dialog.
Select the source and target location(s).
Select the tables that need to be refreshed.
Select the level of parallelism for the tables.
Select the Online Refresh option and set Skip Previous Integrate. This is because we assume two integrate locations.
Select the Schedule box and choose a sensible name for the refresh task.
Click Schedule.
Suspend the integrate jobs. You can leave the capture jobs running.
(Optionally) Go to the Scheduler, select the refresh job that has just been created, and choose New Attribute.
Create a
retry_max
attribute and set this to the desired value. 0 means no retry.Right-click the refresh job and click Start to run the job.