The following error is observed when activating a config to remove a table from replication:
05/16/02 09:24 Error: sp_tconf (for o. SID_name queue *) 14017 - object OWNER.TABLENAME is a(n) UNKNOWN object.
05/16/02 09:24 Error: sp_tconf (for o. SID_name queue *) Failed to lock table OWNER. TABLENAME due to: ORA-00942: table or view does not exist.
05/16/02 09:24 Config compilation completed: datasrc - o. SID_name; file - config_filename
05/16/02 09:24 Bad config file: file - config_filename; Error - LockTable failed, sp_tconf exit
Replicating tables had been previously dropped from the database. Any attempt to remove them from replication after they have been dropped would result in the error since Shareplex cannot find them in the database anymore. When removing a table from replication Shareplex attempts to lock them (to rule out any open transactions) and since they have been dropped, it is not able to find them.
To get around the current problem, recreate the tables which were dropped, then activate the config, and then drop the tables after activation completes.
To avoid this problem in the future, first remove a table from replication and only afterwords drop the source replicating tables from the database (not the other way around).
When removing tables from replication, Shareplex only refers to the object name and not object id so this workaround will be successful.
