Execute the following steps to implement this:
1. Stop the Post queue cleanly on Target.
sp_ctrl> stop post
2. Issue flush from Source to establish a reference point:
sp_ctrl> flush o.SID -- where SID denotes the Source Oracle SID.
3. Make a copy of the active config file.
sp_ctrl> copy config <existingconfig> to <newconfig>
Edit the new copied config file.
sp_ctrl> edit config <newconfig>
Add or edit a new route entry for the problematic table containing the new named Post queue.
Save the new config file.
4. Verify the new config file.
sp_ctrl> verify config <newconfig>
5. Activate the new config file.
sp_ctrl> activate config <newconfig>
6. Assuming that Capture is current with Oracle, the new Post queue will be immediately be spawned on Target but will remain in a stopped state.
If Capture is not current, it may be spawned once Capture is current with Oracle.
7. Start the existing Post queue (which has the problematic table as well as other tables) and let the existing messages drain.
sp_ctrl> start post queue <existingpostqueue>
Once the existing Post queue has drained, it will stop due to the previously executed flush.
At this point, this existing Post queue will not have any messages for the problematic table.
You can then start both Post queues manually.
sp_ctrl> start post
