To create database tables using the Database Initializer:
Start up your target database. You should have already configured the JDBC driver and created any necessary accounts and tablespaces. See the documentation for your database management system.
Start up the database you are currently running ATG applications against (such as SOLID).
Use the application assembler to create an EAR file that includes the
ATGDBSetupmodule, along with all of the modules that you want to create tables for. See Assembling Applications in the ATG Programming Guide.Deploy your EAR file on the application server.
Start the application server.
Point your browser at the start page of the Database Initializer:
http://hostname:port/ATGDBSetupRun the Database Initializer by clicking the Start Initializer button. Follow the instructions in the wizard for specifying the
DataSourcefor connecting to the target database, selecting various options for running the table creation job, and specifying the set of ATG modules to run scripts for. When you reach the Confirm Job Start page, read the page carefully to verify that you want to proceed with the job. If everything looks correct, click the Start Database Job button to create the database tables.Reconfigure your ATG data sources to point to the data sources for your application server. See Configuring the Data Sources and Transaction Manager.
Specifying a Data Source
When you start the Database Initializer, the first step is to specify a data source that points to the target database, which is generally the production-quality database that your application server is already using. The Database Initializer communicates with the database through this data source when it runs the table creation scripts.
The Database Initializer gives you a number of options for specifying this data source:
A list of the data sources currently used by the application server.
A list of Nucleus components that are data sources, plus a text field where you can manually enter the full pathname of a Nucleus component.
A text field for specifying the JNDI name for a data source.
Note, however, that specifying this data source does not change the configuration of your ATG application. Resources in the application that require persistent storage, such as repositories and JMS providers, continue to use the database you started the application against (typically SOLID). Once you have created the tables in the target database, you need to reconfigure your application to use the application server data sources, as described in Configuring the Data Sources and Transaction Manager.
Avoiding Table Conflicts
The Database Initializer gives you two options for specifying the set of modules to run scripts for. You can either run scripts for all currently running modules, or you can choose the modules from a list of all installed ATG modules.
The safest approach is to run the scripts for all currently running modules. This ensures that there are no incompatibilities between any of the modules. If you select this option, however, you must be sure to assemble your EAR file with all of the modules you want to create tables for.
The other option is to select the modules from a list of all modules in your ATG installation. If you use this approach, though, you must be careful not to run multiple scripts that create the same tables. For example, the scripts for the DCS and DCS.Versioned modules create some of the same tables, so you should not run scripts for both of these modules. The Database Initializer will not prevent this, but your database will produce errors when the scripts are run. To prevent conflicts, you should avoid selecting modules that ATG applications cannot run simultaneously.
Importing Data
For most ATG modules, the Database Initializer lists only the scripts for creating the database tables. The tables themselves are all that is needed to start up those modules against the production database.
Some modules, however, also require their tables to be prepopulated with certain data. Therefore, these modules also list repository template files for loading this data. For each template file that you select, the Database Initializer prompts you to specify the corresponding Nucleus repository component.
The following table shows the names of these repository template files and their corresponding Nucleus repository components:
Portal.paf Module | |
|---|---|
|
|
Publishing.base Module | |
|
|
|
|
|
|
BIZUI Module | |
|
|
|
|
|
|
Note: To load a repository template file, the repository needs to be running. Therefore, you should select these files only if the corresponding modules are currently running. The easiest way to ensure this is to run the Database Initializer using the option that creates tables and data for only the currently running modules.
