Skip Headers
Oracle® Streams Replication Administrator's Guide
11g Release 1 (11.1)

Part Number B28322-01
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
View PDF

8 Adding to an Oracle Streams Replication Environment

This chapter contains instructions for adding database objects and databases to an existing Oracle Streams replication environment.

This chapter contains these topics:

Note:

See Also:

Adding Shared Objects to an Existing Single-Source Environment

You add existing database objects to an existing single-source environment by adding the necessary rules to the appropriate capture processes, synchronous captures, propagations, and apply processes. Before creating or altering capture or propagation rules in a running Oracle Streams environment, ensure that any propagations or apply processes that will receive LCRs as a result of the new or altered rules are configured to handle these LCRs. That is, the propagations or apply processes should exist, and each one should be associated with rule sets that handle the LCRs appropriately. If these propagations and apply processes are not configured properly to handle these LCRs, then LCRs can be lost.

For example, suppose you want to add a table to an Oracle Streams environment that already captures, propagates, and applies changes to other tables. Assume only one capture process or synchronous captures will capture changes to this table, and only one apply process will apply changes to this table. In this case, you must add one or more table rules to the following rule sets:

If you perform administrative steps in the wrong order, you can lose LCRs. For example, if you add the rule to a capture process rule set first, without stopping the capture process, then the propagation will not propagate the changes if it does not have a rule that instructs it to do so, and the changes can be lost.

This example assumes that the shared database objects are read-only at the destination databases. If the shared objects are read/write at the destination databases, then the replication environment will not stay synchronized because Oracle Streams is not configured to replicate the changes made to the shared objects at the destination databases.

Figure 8-1 shows the additional configuration steps that must be completed to add shared database objects to a single-source Oracle Streams environment.

Figure 8-1 Example of Adding Shared Objects to a Single-Source Environment

Description of Figure 8-1 follows
Description of "Figure 8-1 Example of Adding Shared Objects to a Single-Source Environment"

To avoid losing LCRs, you should complete the configuration in the following order:

  1. At each source database where shared objects are being added, specify supplemental logging for the added shared objects. See "Managing Supplemental Logging in an Oracle Streams Replication Environment" for instructions.

  2. Either stop the capture process, one of the propagations, or the apply processes:

    • Use the STOP_CAPTURE procedure in the DBMS_CAPTURE_ADM package to stop a capture process.

    • Use the STOP_PROPAGATION procedure in the DBMS_PROPAGATION_ADM package to stop a propagation.

    • Use the STOP_APPLY procedure in the DBMS_APPLY_ADM package to stop an apply process.

    In general, it is best to stop the capture process so that messages do not accumulate in queues during the operation.

    Note:

    Synchronous captures cannot be stopped.

    See Also:

    Oracle Streams Concepts and Administration for more information about completing these tasks
  3. Add the relevant rules to the rule sets for the apply processes. To add rules to the rule set for an apply process, you can run one of the following procedures:

    Excluding the ADD_SUBSET_RULES procedure, these procedures can add rules to the positive or negative rule set for an apply process. The ADD_SUBSET_RULES procedure can add rules only to the positive rule set for an apply process.

  4. Add the relevant rules to the rule sets for the propagations. To add rules to the rule set for a propagation, you can run one of the following procedures:

    Excluding the ADD_SUBSET_PROPAGATION_RULES procedure, these procedures can add rules to the positive or negative rule set for a propagation. The ADD_SUBSET_PROPAGATION_RULES procedure can add rules only to the positive rule set for a propagation.

  5. Add the relevant rules to the rule sets used by the capture process or synchronous capture. To add rules to a rule set for an existing capture process, you can run one of the following procedures and specify the existing capture process:

    Excluding the ADD_SUBSET_RULES procedure, these procedures can add rules to the positive or negative rule set for a capture process. The ADD_SUBSET_RULES procedure can add rules only to the positive rule set for a capture process.

    To add rules to a rule set for an existing synchronous capture, you can run one of the following procedures and specify the existing synchronous capture:

    When you a procedure in the DBMS_STREAMS_ADM package to add the capture process rules, it automatically runs the PREPARE_TABLE_INSTANTIATION, PREPARE_SCHEMA_INSTANTIATION, or PREPARE_GLOBAL_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table, specified schema, or entire database, respectively, if the capture process is a local capture process or a downstream capture process with a database link to the source database.

    You must run the appropriate procedure to prepare for instantiation manually if any of the following conditions is true:

    • You use DBMS_RULE_ADM to create or modify rules in a capture process rule set.

    • You do not add rules for the added objects to a capture process rule set, because the capture process already captures changes to these objects. In this case, rules for the objects can be added to propagations and apply processes in the environment, but not to the capture process.

    • You use a downstream capture process with no database link to the source database.

    If you must prepare for instantiation manually, then see "Preparing Database Objects for Instantiation at a Source Database" for instructions.

    When you use a procedure in the DBMS_STREAMS_ADM package to add the synchronous capture rules, it automatically runs the PREPARE_SYNC_INSTANTIATION function in the DBMS_CAPTURE_ADM package for the specified table.

  6. At each destination database, either instantiate, or set the instantiation SCN for, each database object you are adding to the Oracle Streams environment. If the database objects do not exist at a destination database, then instantiate them using export/import, transportable tablespaces, or RMAN. If the database objects already exist at a destination database, then set the instantiation SCNs for them manually.

    • To instantiate database objects using export/import, first export them at the source database. Next, import them at the destination database. See "Setting Instantiation SCNs Using Export/Import" for information. Also, see "Instantiating Objects in an Oracle Streams Replication Environment" for information about instantiating objects using export/import, transportable tablespaces, and RMAN.

      Do not allow any changes to the database objects being exported while exporting these database objects at the source database. Do not allow changes to the database objects being imported while importing these database objects at the destination database.

      You can specify a more stringent degree of consistency by using an export parameter such as FLASHBACK_SCN or FLASHBACK_TIME.

    • To set the instantiation SCN for a table, schema, or database manually, run the appropriate procedure or procedures in the DBMS_APPLY_ADM package at a destination database:

      • SET_TABLE_INSTANTIATION_SCN

      • SET_SCHEMA_INSTANTIATION_SCN

      • SET_GLOBAL_INSTANTIATION_SCN

      When you run one of these procedures at a destination database, you must ensure that every added object at the destination database is consistent with the source database as of the instantiation SCN.

      If you run SET_GLOBAL_INSTANTIATION_SCN at a destination database, then set the recursive parameter for this procedure to TRUE so that the instantiation SCN also is set for each schema at the destination database and for the tables owned by these schemas.

      If you run SET_SCHEMA_INSTANTIATION_SCN at a destination database, then set the recursive parameter for this procedure to TRUE so that the instantiation SCN also is set for each table in the schema.

      If you set the recursive parameter to TRUE in the SET_GLOBAL_INSTANTIATION_SCN procedure or the SET_SCHEMA_INSTANTIATION_SCN procedure, then a database link from the destination database to the source database is required. This database link must have the same name as the global name of the source database and must be accessible to the user who executes the procedure. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

      Alternatively, you can perform a metadata export/import to set the instantiation SCNs for existing database objects. If you choose this option, then ensure that no rows are imported. Also, ensure that every added object at the importing destination database is consistent with the source database that performed the export at the time of the export. If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

  7. Start any Oracle Streams client you stopped in Step 2:

    • Use the START_CAPTURE procedure in the DBMS_CAPTURE_ADM package to start a capture process.

    • Use the START_PROPAGATION procedure in the DBMS_PROPAGATION_ADM package start a propagation.

    • Use the START_APPLY procedure in the DBMS_APPLY_ADM package to start an apply process.

    See Also:

    Oracle Streams Concepts and Administration for more information about completing these tasks

You must stop the capture process, disable one of the propagation jobs, or stop the apply process in Step 2 to ensure that the table or schema is instantiated before the first LCR resulting from the added rule(s) reaches the apply process. Otherwise, LCRs could be lost or could result in apply errors, depending on whether the apply process rule(s) have been added.

If you are certain that the added table is not being modified at the source database during this procedure, and that there are no LCRs for the table already in the stream or waiting to be captured, then you can perform Step 7 before Step 6 to reduce the amount of time that an Oracle Streams process or propagation job is stopped.

See Also:

"Add Objects to an Existing Oracle Streams Replication Environment" for a detailed example that adds objects to an existing single-source environment

Adding a New Destination Database to a Single-Source Environment

You add a destination database to an existing single-source environment by creating one or more new apply processes at the new destination database and, if necessary, configuring one or more propagations to propagate changes to the new destination database. You might also need to add rules to existing propagations in the stream that propagates to the new destination database.

As in the example that describes "Adding Shared Objects to an Existing Single-Source Environment", before creating or altering propagation rules in a running Oracle Streams environment, ensure that any propagations or apply processes that will receive LCRs as a result of the new or altered rules are configured to handle these LCRs. Otherwise, LCRs can be lost.

This example assumes that the shared database objects are read-only at the destination databases. If the shared objects are read/write at the destination databases, then the replication environment will not stay synchronized because Oracle Streams is not configured to replicate the changes made to the shared objects at the destination databases.

Figure 8-2 shows the additional configuration steps that must be completed to add a destination database to a single-source Oracle Streams environment.

Figure 8-2 Example of Adding a Destination to a Single-Source Environment

Description of Figure 8-2 follows
Description of "Figure 8-2 Example of Adding a Destination to a Single-Source Environment"

To avoid losing LCRs, you should complete the configuration in the following order:

  1. Complete the necessary tasks to prepare each database in your environment for Oracle Streams:

    • Configure an Oracle Streams administrator.

    • Set initialization parameters relevant to Oracle Streams.

    • Configure network connectivity and database links.

    Some of these tasks might not be required at certain databases.

    See Also:

    Oracle Streams Concepts and Administration for more information about preparing a database for Oracle Streams
  2. Create any necessary ANYDATA queues that do not already exist at the destination database. When you create an apply process, you associate the apply process with a specific ANYDATA queue. See "Creating an ANYDATA Queue to Stage LCRs" for instructions.

  3. Create one or more apply processes at the new destination database to apply the changes from its source database. Ensure that each apply process uses rule sets that are appropriate for applying changes. Do not start any of the apply processes at the new database. See "Creating an Apply Process That Applies Captured LCRs" for instructions.

    Keeping the apply processes stopped prevents changes made at the source databases from being applied before the instantiation of the new database is completed, which would otherwise lead to incorrect data and errors.

  4. Configure any necessary propagations to propagate changes from the source databases to the new destination database. Ensure that each propagation uses rule sets that are appropriate for propagating changes. See "Creating a Propagation that Propagates LCRs".

  5. At the source database, prepare for instantiation each database object for which changes will be applied by an apply process at the new destination database.

    If you are using one or more capture processes, then run either the PREPARE_TABLE_INSTANTIATION, PREPARE_SCHEMA_INSTANTIATION, or PREPARE_GLOBAL_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table, specified schema, or entire database, respectively.

    If you are using one or more synchronous captures, then run the PREPARE_SYNC_INSTANTIATION function in the DBMS_CAPTURE_ADM package for the specified table.

    See "Preparing Database Objects for Instantiation at a Source Database" for instructions.

  6. At the new destination database, either instantiate, or set the instantiation SCNs for, each database object for which changes will be applied by an apply process. If the database objects do not already exist at the new destination database, then instantiate them using export/import, transportable tablespaces, or RMAN. If the database objects exist at the new destination database, then set the instantiation SCNs for them.

    • To instantiate database objects using export/import, first export them at the source database. Next, import them at the destination database. See "Setting Instantiation SCNs Using Export/Import" for information. Also, see "Instantiating Objects in an Oracle Streams Replication Environment" for information about instantiating objects using export/import, transportable tablespaces, and RMAN.

      Do not allow any changes to the database objects being exported while exporting these database objects at the source database. Do not allow changes to the database objects being imported while importing these database objects at the destination database.

      You can specify a more stringent degree of consistency by using an export parameter such as FLASHBACK_SCN or FLASHBACK_TIME.

    • To set the instantiation SCN for a table, schema, or database manually, run the appropriate procedure or procedures in the DBMS_APPLY_ADM package at the new destination database:

      • SET_TABLE_INSTANTIATION_SCN

      • SET_SCHEMA_INSTANTIATION_SCN

      • SET_GLOBAL_INSTANTIATION_SCN

      When you run one of these procedures, you must ensure that the shared objects at the new destination database are consistent with the source database as of the instantiation SCN.

      If you run SET_GLOBAL_INSTANTIATION_SCN at a destination database, then set the recursive parameter for this procedure to TRUE so that the instantiation SCN also is set for each schema at the destination database and for the tables owned by these schemas.

      If you run SET_SCHEMA_INSTANTIATION_SCN at a destination database, then set the recursive parameter for this procedure to TRUE so that the instantiation SCN also is set for each table in the schema.

      If you set the recursive parameter to TRUE in the SET_GLOBAL_INSTANTIATION_SCN procedure or the SET_SCHEMA_INSTANTIATION_SCN procedure, then a database link from the destination database to the source database is required. This database link must have the same name as the global name of the source database and must be accessible to the user who executes the procedure. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

      Alternatively, you can perform a metadata export/import to set the instantiation SCNs for existing database objects. If you choose this option, then ensure that no rows are imported. Also, ensure that the shared objects at the importing destination database are consistent with the source database that performed the export at the time of the export. If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

  7. Start the apply processes you created in Step 3 using the START_APPLY procedure in the DBMS_APPLY_ADM package.

See Also:

"Add a Database to an Existing Oracle Streams Replication Environment" for detailed example that adds a database to an existing single-source environment

Adding Shared Objects to an Existing Multiple-Source Environment

You add existing database objects to an existing multiple-source environment by adding the necessary rules to the appropriate capture processes, synchronous captures, propagations, and apply processes.

This example uses the following terms:

Before creating or altering capture or propagation rules in a running Oracle Streams environment, ensure that any propagations or apply processes that will receive LCRs as a result of the new or altered rules are configured to handle these LCRs. That is, the propagations or apply processes should exist, and each one should be associated with rule sets that handle the LCRs appropriately. If these propagations and apply processes are not configured properly to handle these LCRs, then LCRs can be lost.

For example, suppose you want to add a new table to an Oracle Streams environment that already captures, propagates, and applies changes to other tables. Assume multiple capture processes or synchronous captures in the environment will capture changes to this table, and multiple apply processes will apply changes to this table. In this case, you must add one or more table rules to the following rule sets:

If you perform administrative steps in the wrong order, you can lose LCRs. For example, if you add the rule to a capture process rule set first, without stopping the capture process, then the propagation will not propagate the changes if it does not have a rule that instructs it to do so, and the changes can be lost.

Figure 8-3 shows the additional configuration steps that must be completed to add shared database objects to a multiple-source Oracle Streams environment.

Figure 8-3 Example of Adding Shared Objects to a Multiple-Source Environment

Description of Figure 8-3 follows
Description of "Figure 8-3 Example of Adding Shared Objects to a Multiple-Source Environment"

When there are multiple source databases in an Oracle Streams replication environment, change cycling is possible. Change cycling happens when a change is sent back to the database where it originated. Typically, you should avoid change cycling. Before you configure your replication environment, see Chapter 4, "Oracle Streams Tags", and ensure that you configure the replication environment to avoid change cycling.

To avoid losing LCRs, you should complete the configuration in the following order:

  1. At each populated database, specify any necessary supplemental logging for the objects being added to the environment. See "Managing Supplemental Logging in an Oracle Streams Replication Environment" for instructions.

  2. Either stop all of the capture processes that will capture changes to the added objects, stop all of the propagations that will propagate changes to the added objects, or stop all of the apply process that will apply changes to the added objects:

    • Use the STOP_CAPTURE procedure in the DBMS_CAPTURE_ADM package to stop a capture process.

    • Use the STOP_PROPAGATION procedure in the DBMS_PROPAGATION_ADM package to stop a propagation.

    • Use the STOP_APPLY procedure in the DBMS_APPLY_ADM package to stop an apply process.

    In general, it is best to stop the capture process so that messages do not accumulate in queues during the operation.

    Note:

    Synchronous captures cannot be stopped.

    See Also:

    Oracle Streams Concepts and Administration for more information about completing these tasks
  3. Add the relevant rules to the rule sets for the apply processes that will apply changes to the added objects. To add rules to the rule set for an apply process, you can run one of the following procedures:

    Excluding the ADD_SUBSET_RULES procedure, these procedures can add rules to the positive or negative rule set for an apply process. The ADD_SUBSET_RULES procedure can add rules only to the positive rule set for an apply process.

  4. Add the relevant rules to the rule sets for the propagations that will propagate changes to the added objects. To add rules to the rule set for a propagation, you can run one of the following procedures:

    Excluding the ADD_SUBSET_PROPAGATION_RULES procedure, these procedures can add rules to the positive or negative rule set for a propagation. The ADD_SUBSET_PROPAGATION_RULES procedure can add rules only to the positive rule set for a propagation.

  5. Add the relevant rules to the rule sets used by each capture process or synchronous capture that will capture changes to the added objects. To add rules to a rule set for an existing capture process, you can run one of the following procedures and specify the existing capture process:

    Excluding the ADD_SUBSET_RULES procedure, these procedures can add rules to the positive or negative rule set for a capture process. The ADD_SUBSET_RULES procedure can add rules only to the positive rule set for a capture process.

    To add rules to a rule set for an existing synchronous capture, you can run one of the following procedures and specify the existing synchronous capture:

    When you a procedure in the DBMS_STREAMS_ADM package to add the capture process rules, it automatically runs the PREPARE_TABLE_INSTANTIATION, PREPARE_SCHEMA_INSTANTIATION, or PREPARE_GLOBAL_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table, specified schema, or entire database, respectively, if the capture process is a local capture process or a downstream capture process with a database link to the source database.

    You must run the appropriate procedure to prepare for instantiation manually if any of the following conditions is true:

    • You use DBMS_RULE_ADM to create or modify rules in a capture process rule set.

    • You do not add rules for the added objects to a capture process rule set, because the capture process already captures changes to these objects. In this case, rules for the objects can be added to propagations and apply processes in the environment, but not to the capture process.

    • You use a downstream capture process with no database link to the source database.

    If you must prepare for instantiation manually, then see "Preparing Database Objects for Instantiation at a Source Database" for instructions.

    When you use a procedure in the DBMS_STREAMS_ADM package to add the synchronous capture rules, it automatically runs the PREPARE_SYNC_INSTANTIATION function in the DBMS_CAPTURE_ADM package for the specified table.

After completing these steps, complete the steps in each of the following sections that apply to your environment. You might need to complete the steps in only one of these sections or in both of these sections:

Configuring Populated Databases When Adding Shared Objects

After completing the steps in "Adding Shared Objects to an Existing Multiple-Source Environment", complete the following steps for each populated database if your environment has more than one populated database:

  1. For each populated database, set the instantiation SCN for each added object at the other populated databases in the environment. These instantiation SCNs must be set, and only the changes made at a particular populated database that are committed after the corresponding SCN for that database will be applied at another populated database.

    For each populated database, you can set these instantiation SCNs for each added object in one of the following ways:

    1. Perform a metadata only export of the added objects at the populated database and import the metadata at each of the other populated databases. Such an import sets the required instantiation SCNs for the database at the other databases. Ensure that no rows are imported. Also, ensure that the shared objects at each of the other populated databases are consistent with the populated database that performed the export at the time of the export.

      If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

    2. Set the instantiation SCNs manually for the added objects at each of the other populated databases. Ensure that every added object at each populated database is consistent with the instantiation SCNs you set at that database. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

Adding Shared Objects to Import Databases in an Existing Environment

After completing the steps in "Adding Shared Objects to an Existing Multiple-Source Environment", complete the following steps for the import databases:

  1. Pick the populated database that you will use as the export database. Do not perform the instantiations yet.

  2. For each import database, set the instantiation SCNs for the added objects at all of the other databases in the environment that will be a destination database of the import database. In this case, the import database will be the source database for these destination databases. The databases where you set the instantiation SCNs might be populated databases and other import databases.

    1. If one or more schemas will be created at an import database during instantiation or by a subsequent shared DDL change, then run the SET_GLOBAL_INSTANTIATION_SCN procedure in the DBMS_APPLY_ADM package for this import database at all of the other databases in the environment.

    2. If a schema exists at an import database, and one or more tables will be created in the schema during instantiation or by a subsequent shared DDL change, then run the SET_SCHEMA_INSTANTIATION_SCN procedure in the DBMS_APPLY_ADM package for the schema for this import database at each of the other databases in the environment. Do this for each such schema.

    See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

    Because you are running these procedures before any tables are instantiated at the import databases, and because the local capture processes or synchronous captures are configured already for these import databases, you will not need to run the SET_TABLE_INSTANTIATION_SCN procedure for each table created during instantiation. Instantiation SCNs will be set automatically for these tables at all of the other databases in the environment that will be destination databases of the import database.

  3. At the export database you chose in Step 1, perform an export of the shared objects. Next, perform an import of the shared objects at each import database. See "Instantiating Objects in an Oracle Streams Replication Environment" and Oracle Database Utilities for information about using export/import.

    Do not allow any changes to the database objects being exported while exporting these database objects at the source database. Do not allow changes to the database objects being imported while importing these database objects at the destination database.

    You can specify a more stringent degree of consistency by using an export parameter such as FLASHBACK_SCN or FLASHBACK_TIME.

  4. For each populated database, except for the export database, set the instantiation SCNs for the added objects at each import database that will be a destination database of the populated source database. These instantiation SCNs must be set, and only the changes made at a populated database that are committed after the corresponding SCN for that database will be applied at an import database.

    For each populated database, you can set these instantiation SCNs for the added objects in one of the following ways:

    1. Perform a metadata only export of the added objects at the populated database and import the metadata at each import database. Each import sets the required instantiation SCNs for the populated database at the import database. In this case, ensure that every added object at the import database is consistent with the populated database at the time of the export.

      If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

    2. Set the instantiation SCNs manually for the added objects at each import database. Ensure that every added object at each import database is consistent with the populated database as of the corresponding instantiation SCN. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

Complete the Adding Objects to a Multiple-Source Environment Configuration

Before completing the configuration, you should have completed the following tasks:

When all of the previous configuration steps are finished, complete the following steps:

  1. At each database, configure conflict resolution for the added database objects if conflicts are possible. See "Managing Oracle Streams Conflict Detection and Resolution" for instructions.

  2. Start each Oracle Streams client you stopped in Step 2 in "Adding Shared Objects to an Existing Multiple-Source Environment":

    • Use the START_CAPTURE procedure in the DBMS_CAPTURE_ADM package to start a capture process.

    • Use the START_PROPAGATION procedure in the DBMS_PROPAGATION_ADM package to start a propagation.

    • Use the START_APPLY procedure in the DBMS_APPLY_ADM package to start an apply process.

    See Also:

    Oracle Streams Concepts and Administration for more information about completing these tasks

Adding a New Database to an Existing Multiple-Source Environment

Figure 8-4 shows the additional configuration steps that must be completed to add a source/destination database to a multiple-source Oracle Streams environment.

Figure 8-4 Example of Adding a Database to a Multiple-Source Environment

Description of Figure 8-4 follows
Description of "Figure 8-4 Example of Adding a Database to a Multiple-Source Environment"

When there are multiple source databases in an Oracle Streams replication environment, change cycling is possible. Change cycling happens when a change is sent back to the database where it originated. Typically, you should avoid change cycling. Before you configure your replication environment, see Chapter 4, "Oracle Streams Tags", and ensure that you configure the replication environment to avoid change cycling.

Complete the following steps to add a new source/destination database to an existing multiple-source Oracle Streams environment:

Note:

Ensure that no changes are made to the objects being shared at the database you are adding to the Oracle Streams environment until the instantiation at the database is complete.
  1. Complete the necessary tasks to prepare each database in your environment for Oracle Streams:

    • Configure an Oracle Streams administrator.

    • Set initialization parameters relevant to Oracle Streams.

    • For each database that will run a capture process, prepare the database to run a capture process.

    • Configure network connectivity and database links.

    Some of these tasks might not be required at certain databases.

    See Also:

    Oracle Streams Concepts and Administration for more information about preparing a database for Oracle Streams
  2. Create any necessary ANYDATA queues that do not already exist. When you create a capture process, synchronous capture, or apply process, you associate the process with a specific ANYDATA queue. When you create a propagation, you associate it with a specific source queue and destination queue. See "Creating an ANYDATA Queue to Stage LCRs" for instructions.

  3. Create one or more apply processes at the new database to apply the changes from its source databases. Ensure that each apply process uses rule sets that are appropriate for applying changes. Do not start any apply process at the new database. See "Creating an Apply Process That Applies Captured LCRs" for instructions.

    Keeping the apply processes stopped prevents changes made at the source databases from being applied before the instantiation of the new database is completed, which would otherwise lead to incorrect data and errors.

  4. If the new database will be a source database, then, at all databases that will be destination databases for the changes made at the new database, create one or more apply processes to apply changes from the new database. Ensure that each apply process uses rule sets that are appropriate for applying changes. Do not start any of these new apply processes. See "Creating an Apply Process That Applies Captured LCRs" for instructions.

  5. Configure propagations at the databases that will be source databases of the new database to send changes to the new database. Ensure that each propagation uses rule sets that are appropriate for propagating changes. See "Creating a Propagation that Propagates LCRs".

  6. If the new database will be a source database, then configure propagations at the new database to send changes from the new database to each of its destination databases. Ensure that each propagation uses rule sets that are appropriate for propagating changes. See "Creating a Propagation that Propagates LCRs".

  7. If the new database will be a source database, and the shared objects already exist at the new database, then specify any necessary supplemental logging for the shared objects at the new database. See "Managing Supplemental Logging in an Oracle Streams Replication Environment" for instructions.

  8. At each source database for the new database, prepare for instantiation each database object for which changes will be applied by an apply process at the new database.

    If you are using one or more capture processes, then run either the PREPARE_TABLE_INSTANTIATION, PREPARE_SCHEMA_INSTANTIATION, or PREPARE_GLOBAL_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table, specified schema, or entire database, respectively.

    If you are using one or more synchronous captures, then run the PREPARE_TABLE_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table.

    See "Preparing Database Objects for Instantiation at a Source Database" for instructions.

  9. If the new database will be a source database, then create one or more capture processes or synchronous captures to capture the relevant changes. See "Creating a Capture Process" for instructions. If you plan to use capture processes, then Oracle recommends that you use only one capture process for each source database.

    When you use a procedure in the DBMS_STREAMS_ADM package to add the capture process rules, it automatically runs the PREPARE_TABLE_INSTANTIATION, PREPARE_SCHEMA_INSTANTIATION, or PREPARE_GLOBAL_INSTANTIATION procedure in the DBMS_CAPTURE_ADM package for the specified table, specified schema, or entire database, respectively, if the capture process is a local capture process or a downstream capture process with a database link to the source database.

    You must run the appropriate procedure to prepare for instantiation manually if any of the following conditions is true:

    • You use the DBMS_RULE_ADM package to add or modify rules.

    • You use an existing capture process and do not add capture process rules for any shared object.

    • You use a downstream capture process with no database link to the source database.

    If you must prepare for instantiation manually, then see "Preparing Database Objects for Instantiation at a Source Database" for instructions.

    When you use a procedure in the DBMS_STREAMS_ADM package to add the synchronous capture rules, it automatically runs the PREPARE_SYNC_INSTANTIATION function in the DBMS_CAPTURE_ADM package for the specified table.

  10. If the new database will be a source database, then start any capture process you created in Step 9 using the START_CAPTURE procedure in the DBMS_CAPTURE_ADM package.

After completing these steps, complete the steps in the appropriate section:

Configuring Databases If the Shared Objects Already Exist at the New Database

After completing the steps in "Adding a New Database to an Existing Multiple-Source Environment", complete the following steps if the objects that are to be shared with the new database already exist at the new database:

  1. For each source database of the new database, set the instantiation SCNs at the new database. These instantiation SCNs must be set, and only the changes made at a source database that are committed after the corresponding SCN for that database will be applied at the new database.

    For each source database of the new database, you can set these instantiation SCNs in one of the following ways:

    1. Perform a metadata only export of the shared objects at the source database and import the metadata at the new database. The import sets the required instantiation SCNs for the source database at the new database. Ensure that no rows are imported. In this case, ensure that the shared objects at the new database are consistent with the source database at the time of the export.

      If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

    2. Set the instantiation SCNs manually at the new database for the shared objects. Ensure that the shared objects at the new database are consistent with the source database as of the corresponding instantiation SCN. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

  2. For the new database, set the instantiation SCNs at each destination database of the new database. These instantiation SCNs must be set, and only the changes made at the new source database that are committed after the corresponding SCN will be applied at a destination database. If the new database is not a source database, then do not complete this step.

    You can set these instantiation SCNs for the new database in one of the following ways:

    1. Perform a metadata only export at the new database and import the metadata at each destination database. Ensure that no rows are imported. The import sets the required instantiation SCNs for the new database at each destination database. In this case, ensure that the shared objects at each destination database are consistent with the new database at the time of the export.

      If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

    2. Set the instantiation SCNs manually at each destination database for the shared objects. Ensure that the shared objects at each destination database are consistent with the new database as of the corresponding instantiation SCN. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

  3. At the new database, configure conflict resolution if conflicts are possible. See "Managing Oracle Streams Conflict Detection and Resolution" for instructions.

  4. Start the apply processes that you created at the new database in Step 3 using the START_APPLY procedure in the DBMS_APPLY_ADM package.

  5. Start the apply processes that you created at each of the other destination databases in Step 4. If the new database is not a source database, then do not complete this step.

Adding Shared Objects to a New Database

After completing the steps in "Adding a New Database to an Existing Multiple-Source Environment", complete the following steps if the objects that are to be shared with the new database do not already exist at the new database:

  1. If the new database is a source database for other databases, then, at each destination database of the new source database, set the instantiation SCNs for the new database.

    1. If one or more schemas will be created at the new database during instantiation or by a subsequent shared DDL change, then run the SET_GLOBAL_INSTANTIATION_SCN procedure in the DBMS_APPLY_ADM package for the new database at each destination database of the new database.

    2. If a schema exists at the new database, and one or more tables will be created in the schema during instantiation or by a subsequent shared DDL change, then run the SET_SCHEMA_INSTANTIATION_SCN procedure in the DBMS_APPLY_ADM package for the schema at each destination database of the new database. Do this for each such schema.

    See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

    Because you are running these procedures before any tables are instantiated at the new database, and because the local capture process or synchronous capture is configured already at the new database, you will not need to run the SET_TABLE_INSTANTIATION_SCN procedure for each table created during instantiation. Instantiation SCNs will be set automatically for these tables at all of the other databases in the environment that will be destination databases of the new database.

    If the new database will not be a source database, then do not complete this step, and continue with the next step.

  2. Pick one source database from which to instantiate the shared objects at the new database using export/import. First, perform an export of the shared objects. Next, perform an import of the shared objects at the new database. See "Instantiating Objects in an Oracle Streams Replication Environment" and Oracle Database Utilities for information about using export/import.

    Do not allow any changes to the database objects being exported while exporting these database objects at the source database. Do not allow changes to the database objects being imported while importing these database objects at the destination database.

    You can specify a more stringent degree of consistency by using an export parameter such as FLASHBACK_SCN or FLASHBACK_TIME.

  3. For each source database of the new database, except for the source database that performed the export for instantiation in Step 2, set the instantiation SCNs at the new database. These instantiation SCNs must be set, and only the changes made at a source database that are committed after the corresponding SCN for that database will be applied at the new database.

    For each source database, you can set these instantiation SCNs in one of the following ways:

    1. Perform a metadata only export at the source database and import the metadata at the new database. The import sets the required instantiation SCNs for the source database at the new database. In this case, ensure that the shared objects at the new database are consistent with the source database at the time of the export.

      If you are sharing DML changes only, then table level export/import is sufficient. If you are sharing DDL changes also, then additional considerations apply. See "Setting Instantiation SCNs Using Export/Import" for more information about performing a metadata export/import.

    2. Set the instantiation SCNs manually at the new database for the shared objects. Ensure that the shared objects at the new database are consistent with the source database as of the corresponding instantiation SCN. See "Setting Instantiation SCNs Using the DBMS_APPLY_ADM Package" for instructions.

  4. At the new database, configure conflict resolution if conflicts are possible. See "Managing Oracle Streams Conflict Detection and Resolution" for instructions.

  5. Start the apply processes that you created in Step 3 at the new database using the START_APPLY procedure in the DBMS_APPLY_ADM package.

  6. Start the apply processes that you created in Step 4 at each of the other destination databases. If the new database is not a source database, then do not complete this step.