12 Using the Oracle Transparent Gateway for DRDA

Processing a Database Link

The database and application administrators of a distributed database system are responsible for managing the database links that define paths to the DRDA database. The tasks are as follows:

• Creating Database Links

• Dropping Database Links

• Examining Available Database Links

• Limiting the Number of Active Database Links

CREATE PUBLIC DATABASE LINK dblink 
CONNECT TO userid IDENTIFIED BY password  
USING 'tns_name_entry';

DROP PUBLIC DATABASE LINK dblink;

Accessing the Gateway

To access the gateway, complete the following steps on the Oracle integrating server:

1. Step 1: Login to the Oracle Integrating Server

2. Step 2: Creating a database Link to the DRDA Database

3. Step 3: Retrieve data from the DRDA Database

Step 1: Login to the Oracle Integrating Server

Log in to the Oracle Integrating Server to access the gateway

Step 2: Creating a database Link to the DRDA Database

For example, use:

CREATE PUBLIC DATABASE LINK DRDA
CONNECT TO ORADRDA IDENTIFIED BY oracle_pw 
USING 'tns_name_entry' 

Step 3: Retrieve data from the DRDA Database

This query fetches the TABLE file in the library SECURE, using the name ORACLE as the DRDA Server user profile. The ORACLE user profile must have the appropriate privileges on the DRDA Server to access the SECURE.TABLE files:

SELECT * FROM SECURE.TABLE@DRDA

The following is an example of the error messages that are displayed if insufficicent privileges are displayed:

ORA-1031:   insufficient privileges
TG4DRDA V10.2.0.1.0 grc=0, drc=-777 (83TC,0000), errp=ARIXO,
sqlcode=-551, sqlstate=42501, errd=FFFFFF9C,0,0,0,0,0
errmc=USER SELECT SECURE.TABLE

Accessing AS/400 File Members

Nothing specific to DRDA or to the gateway controls the access to AS/400 files and file members. However, DB2/400 uses a naming convention that implies that the file member name is the same as the name of the file being addressed. For example, accessing schema.table implies that table is the file name and also that table is the file member name being accessed.

To access file members with names that differ from the associated file name, you must create a view within the file so that DB2/400 can reference the correct file member.

One method for creating this view involves issuing the console command Create Logical File (CRTLF). This action creates a logical association between the file name and the file member name.

Using the Synonym Feature

You can provide complete data, location, and network transparency by using the synonym feature of Oracle Database. When a synonym is defined, the user need not know the underlying table or network protocol being used. A synonym can be public, which means it is available to all Oracle users. A synonym can also be defined as private, available only to the user who created it. Refer to Oracle Database Reference for details on the synonym feature.

The following statement creates a system-wide synonym for the EMP file in the DRDA Server with ownership of ORACLE:

CREATE PUBLIC SYNONYM EMP FOR ORACLE.EMP@DRDA

Performing Distributed Queries

The Oracle Transparent gateway technology enables the execution of distributed queries that join Oracle Database and DRDA Servers and any other data store for which Oracle provides a gateway. These complex operations can be completely transparent to the users requesting the data.

The distributed query optimizer (DQO) capability can provide better performance of distributed queries. Statistical data regarding tables from DRDA Server is retrieved and passed to the Oracle Integrating Server. The DQO capability is enabled or disabled by the DRDA_OPTIMIZE_QUERY parameter. Refer to "DRDA_OPTIMIZE_QUERY" for more information.

The following example joins data between an Oracle Database, DB2/OS390, and a DRDA Server:

SELECT o.custname, p.projno, e.ename, sum(e.rate*p.hours)
FROM orders@DB2 o, EMP@ORACLE7 e, projects@DRDA p
WHERE o.projno = p.projno 
AND p.empno = e.empno
GROUP BY o.custname, p.projno, e.ename

A combination of views and synonyms, using the following SQL statements, keeps the process of distributed queries transparent to the user:

CREATE SYNONYM orders for orders@DB2;
CREATE SYNONYM PROJECTS for PROJECTS@DRDA;
CREATE VIEW details (custname,projno,ename,spend)
AS 
SELECT o.custname, p.projno, e.ename, sum(e.rate*p.hours)
FROM orders o, EMP e, projects p
WHERE o.projno = p.projno
AND p.empno = e.empno
GROUP BY o.custname, p.projno, e.ename

The following SQL statement retrieves information from these three data stores in one command:

SELECT * FROM DETAILS;

The results of this command are:

CUSTNAME   PROJNO     ENAME      SPEND
---------  ---------  ---------  ---------
ABC Co.    1         Jones      400
ABC Co.    1         Smith      180
XYZ Inc.   2         Jones      400
XYZ Inc.   2         Smith      180

Read-Only Gateway

The read-only option can provide improved performance and security. This improved performance depends on your configuration and parameter selections. A Gateway Initialization Parameter, DRDA_READ_ONLY, controls whether the gateway is enabled in this mode.

If you enable the read-only option, then only queries (SELECT statements) are allowed by the gateway. The capabilities that control whether updates are allowed by the gateway are disabled. These capabilities include INSERT, UPDATE, DELETE and stored-procedure support (pass-through SQL and DB2 stored procedures). Statements attempting to modify records on the DRDA Server are rejected.

Oracle recommends that you should not routinely switch between settings of the DRDA_READ_ONLY parameter. If you need both the update and DRDA_READ_ONLY functionality, then you should create two separate instances of the gateway with different read-only settings.

Replicating in a Heterogeneous Environment

Oracle Transparent Gateway for DRDA provides a number of options for replicating Oracle and non-Oracle data throughout the enterprise.

Oracle Database 10g Triggers

When updates are made to Oracle Database, synchronous copies of Oracle and non-Oracle data can be maintained automatically by using Oracle Database 10g triggers.

Oracle Snapshots

Oracle Transparent Gateway for DRDA can use the Oracle snapshot feature to automatically replicate non-Oracle data into Oracle Database. The complete refresh capability of Oracle snapshot can be used to propagate a complete copy or a subset of the non-Oracle data into Oracle Database at user-defined intervals.

Copying Data from Oracle Database 10g to DRDA Server

The COPY command enables you to copy data from Oracle Database to a DRDA Server database. The Oracle SQL command INSERT is not supported. If you use the INSERT command:

INSERT INTO DRDA_table SELECT * FROM local_table

then the following message is displayed:

ORA-2025: All tables in the SQL statement must be at the remote database

To copy data from your Oracle database to the DRDA Server, use:

COPY FROM username/password@connect_identifier - 
INSERT destination_table -  
USING query 

For example, to select all rows from the local Oracle EMP table, insert them into the EMP table on the DRDA Server, and commit the transaction, use:

COPY FROM scott/tiger@ORACLE -
INSERT scott.EMP@DRDA -
USING SELECT * FROM EMP

The SQL*Plus COPY command supports APPEND, CREATE, INSERT, and REPLACE commands. However, INSERT is the only command supported when copying to the DRDA Server. For more information about the COPY command, refer to SQL*Plus User's Guide and Reference.

Copying Data from DRDA Server to Oracle Database 10g Server

The CREATE TABLE command enables you to copy data from a DRDA Server to Oracle Database. To create a table on your Oracle database and to insert rows from a DRDA Server table, use:

CREATE TABLE table_name 
AS query 

The following example creates the table EMP in your local Oracle Database and inserts the rows from the EMP table on the DRDA Server:

CREATE TABLE EMP
AS SELECT * FROM scott.EMP@DRDA

Alternatively, you can use the SQL*Plus COPY command to copy data from a DRDA Server to Oracle Database. For more information about the COPY command, refer to SQL*Plus User's Guide and Reference.

Tracing SQL Statements

SQL statements issued through the gateway can be changed before reaching the DRDA database. These changes are made to make the format acceptable to the gateway or to make Oracle SQL compatible with DRDA Server SQL. Oracle Integrating Server and the gateway can change the statements depending on the situation.

For various reasons, you might need to assess whether the gateway has altered the statement correctly or whether the statement could be rewritten to improve performance. SQL tracing is a feature that allows you to view the changes made to a SQL statement by the Oracle Integrating Server or the gateway.

SQL tracing reduces gateway performance. Use tracing only while testing and debugging your application. Do not enable SQL tracing when the application is running in a production environment. For more information about enabling SQL tracing, refer to the section on "SQL Tracing and the Gateway" in Chapter 16, "Error Messages, Diagnosis, and Reporting".