The table below lists and describes the changes to tables that require initialization of a Datareplicator and modification of definitions. Note that changes to tables that are not subject to data linkage do not require Datareplicator initialization or modification of definitions.

Table 6-15 Changes to tables that require Datareplicator initialization and modification of definitions

Change to table	Whether Datareplicator initialization is required	Modification of extraction definition	Modification of import definition
Adding a table#	Y	Add the extract statement.	Add the load statement.
Deleting a table	Y	Delete the extract statement.	Delete the load statement.
Deleting a table and then re-creating the table (changing the table ID)	Y	--	--
Adding a non-mapping-key column to the target of linkage	Y	When an asterisk (*) is specified in the definition of the columns that are to be extracted by using the extract statement: There is no need to modify the definition. When column names are specified in the definition of the columns that are to be extracted by using the extract statement: Add the column name.	When an asterisk (*) is specified in the definition of the fields that are to be imported by using the load statement or when the load statement is omitted: There is no need to modify the definition. When field names are specified in the definition of the columns that are to be extracted by using the load statement: Add the field name.
Adding a mapping-key column to the target of linkage	Y	When an asterisk (*) is specified in the definition of the columns that are to be extracted by using the extract statement: Add the column to the key or ukey clause. If the column is to be a transmission condition, add it to the where clause of the send statement. When column names are specified in the definition of the columns that are to be extracted by using the extract statement: Add the column name to the definition of the columns to be extracted and also add it to the key or ukey clause. If the column is to be a transmission condition, add it to the where clause of the send statement.	When an asterisk (*) is specified in the definition of the fields that are to be imported by using the load statement or when the load statement is omitted: There is no need to modify the definition. When field names are specified in the definition of the columns that are to be extracted by using the load statement: Add the field name.
Removing a non-mapping key-column from the target of linkage	Y	When an asterisk (*) is specified in the definition of the columns that are to be extracted by using the extract statement: There is no need to modify the definition. When column names are specified in the definition of the columns that are to be extracted by using the extract statement: Delete the column name.	When an asterisk (*) is specified in the definition of the fields that are to be imported by using the load statement or when the load statement is omitted: There is no need to modify the definition. When field names are specified in the definition of the columns that are to be extracted by using the load statement: Delete the field name.
Removing a mapping key-column from the target of linkage	Y	When an asterisk (*) is specified in the definition of the columns that are to be extracted by using the extract statement: Delete the column name from the key or ukey clause. If the column is used as a transmission condition, also delete it from the where clause of the send statement. When column names are specified in the definition of the columns that are to be extracted by using the extract statement: Delete the column name from the definition of the columns subject to extraction and also delete it from the key or ukey clause. If the column is used as a transmission condition, also delete it from the where clause of the send statement.	When an asterisk (*) is specified in the definition of the fields that are to be imported by using the load statement or when the load statement is omitted: There is no need to modify the definition. When field names are specified in the definition of the columns that are to be extracted by using the load statement: Delete the field name.
Adding an RDAREA to a back-end server subject to linkage	N	--	--
Adding an RDAREA to a back-end server that was not subject to linkage	Y	There is no need to modify the extraction definition. Modify the extraction environment definition as described below, and then initialize the source Datareplicator. Specify in the dsid operand in the extraction environment definition a data linkage identifier (that is unique among all back-end servers) for the back-end server that will be subject to linkage. If true is specified in the extsuppress operand for the back-end server that will be subject to linkage, specify false.	Add the load statement to the import definition for the data linkage identifier that receives update information. Also, specify the following definitions in addition to the import definition: Define in the dsid operand in the import system definition the data linkage identifier that will be receiving update information. Create the import environment definition for the data linkage identifier that will be receiving update information.
Deleting a column that is not subject to linkage	Y	--	--
Renaming a table	Y	Change the table identifier in the extract statement.	Change the table identifier in the load statement.
Renaming a column	Y	When an asterisk (*) is specified in the definition of the columns that are to be extracted by using the extract statement: There is no need to modify the definition. When column names are specified in the definition of the columns that are to be extracted by using the extract statement: Rename the column.	--
Changing a column's data type (column attribute)	Y	If the column is used as a transmission condition, change the condition in the where clause of the send statement.	--
Changing a partition storage condition	N	--	--

Legend:

Y: Datareplicator must be initialized.

N: There is no need to initialize Datareplicator.

--: There is no need to modify definitions.

#

If you are adding tables that will be subject to extraction, check and, if necessary, revise the extinfonum operand value in the extraction system definition. If the extinfonum operand value is smaller than the number of extract statements, an error occurs during execution of the hdeprep command.