Oracle 10g Configuration
In order to get Cascade running you need to provide a database server where all data will be stored. This set of instructions discusses the process of getting started with Cascade using Oracle 10g.
Getting Started
Please download the example Cascade database:
In order to set the database up for Cascade server it is strongly recommended that you have an existing database with UTF-8 enabled character set. This is necessary to take advantage of Cascade Server's multi-language support. It is strongly recommended that you setup a separate tablespace for your Cascade database. You can start the process of the database preparation by creating the tablespace.
Create Tablespace
You will need to create a tablespace called CASCADE where all Cascade Server tables will reside. The data import (described later) will create a user called CASCADE and the CASCADE schema. Here are the steps:
- With Enterprise Manager opened to the desired database, go to "TABLESPACES" and click "Create".
- Fill in "CASCADE" in the Name field and then under "Datafiles" click the "Add" button (Alternately you can avoid adding a datafile by checking "Use bigfile tablespace" check box, but we strongly recommend creating a separate file.)
- While at the Add Datafile screen fill in "CASCADE.DBF" and point the File Directory to the desired location. Under File Size type 30 MB
- Under storage check "Automatically extend datafile when full (AUTOEXTEND)" check box and in the "Increment" field fill 2 MB. Click "Continue."
- Click "OK" to create the tablespace. The generated SQL should look like this:
CREATE SMALLFILE TABLESPACE "CASCADE" DATAFILE '/u01/app/oracle/oracle/product/10.2.0/db_1/oradata/orcl/CASCADE.DBF' SIZE 30M AUTOEXTEND ON NEXT 2M MAXSIZE UNLIMITED LOGGING EXTENT MANAGEMENT LOCAL SEGMENT SPACE MANAGEMENT AUTO
You can also execute the above code in Sql*Plus to accomplish the same result. When finished with the tablespace creation, we can continue with the data import.
Importing the Example Database
- Go to Maintenance -> Import from Export Files.
- Under Directory Object select from the dropdown menu the correct directory where the cascade import script resides. Under File Name type CASCADE.DMP
- Under Import Type select Entire Files and under Host Credentials fill in the OS username and password for the Oracle account and click "Continue."
- On the next screen (Import: Re-Mapping) you don't need to do anything unless you have created a tablespace and/or user different than CASCADE. Click "Next."
- Under Import: Options it is recommended to create a log file of the import. Select a Directory Object and under Log File type the name of the file (e.g. CASCADE.LOG) and click "Next."
- Now the only thing left is to select a name for the import job and a description: under Job Name type the desired name (e.g. cascade_import and note that the name has to be unique) and under Description you can type an optional job description. Click "Next."
- You will see a screen Import: Process In Progress with an hourglass indicating the start of the job.
- After the job is complete you can examine the import log file as well as the size and contents of the newly created CASCADE tablespace (it should contain around 70 tables and should be 34.8Mb)
Configuring the Datasource File
With the database in place, the final step is to configure the oracle-ds.xml file found in the cascade/server/default/deploy/ folder. Please open the file and set the corresponding values according to the instructions in the file. Make sure and set the "jndi-name" to "SystemDS" e.g. <jndi-name>SystemDS</jndi-name>.
Now, open up the two other datasource files mysql-ds.xml and mssql-ds.xml and set the "jndi-name" to a different value for each (e.g. for mysql-ds.xml set it to "SystemDS4" and for mssql-ds.xml set it to "SystemDS5"). Please also fully delete these two folders if they are present:
- cascade\server\default\tmp
- cascade\server\default\work
You can now start Cascade Server and monitor the console log to verify a successful startup. That's it! Let us know if we can help.