...the answer is out there, but hiding from Google :)
In particular, how do I stop the synchronization process for a single remote, wipe out all traces of the synchronization setup for that remote database from both the consolidated and remote databases, and then redefine and restart synchronization?
We don't have an API to kill an active sync on the server side, which I presume is the case you are interested in. At minimal verbosity level (-v) you'll see a message: "The current synchronization is using a connection with ...". This message identifies the RDBMS connection for any given sync. If your RDBMS supports it, you could use the information to kill the connection. For a more programmatic approach, you could use the -vR switch to add the remote ID to all log messages for a sync, and listen in Java or .NET on all log lines, looking for the message above. Then it would just be a matter of what trigger mechanism and/or business logic you would use to kill the RDBMS connection.
As to wiping, we instead recommend using a different remote ID (eg. a new GUID), so you start clean. This approach has the advantage that it can be initiated client-side, in most deployment scenarios.
answered 16 Jul '12, 09:44