13 Configuring Search Features

This chapter guides you through the process of configuring search for Oracle Collaboration Suite.

This chapter contains the following topics:

Default Oracle Ultra Search Instance
Creating an Oracle Ultra Search Database User
Default Oracle Ultra Search Instance
Configuring the Oracle Ultra Search Crawler
Creating a Web Source
Creating a Schedule
Scheduling the Oracle Ultra Search Crawler
Setting Oracle Ultra Search Properties
Making E-mail Searchable
Restarting OC4J_iFS_files
Testing Oracle Ultra Search

13.1 Default Oracle Ultra Search Instance

The Oracle Ultra Search installer creates a default out of the box Oracle Ultra Search instance based on the default Ultra Search test user, so users can test Oracle Ultra Search functionality based on the default instance after installation.

The default instance name is WK_INST. It is created based on the database user WK_TEST. In other words, WK_TEST is the instance administrator for WK_INST. The default user password is WK_TEST.

For security purposes, WK_TEST is locked after the installation. The Oracle Collaboration Suite administrator should login to the database as DBA role, unlock the WK_TEST user account, and set the password to be WK_TEST. (The password expires after the installation.) If the password is changed to anything other than WK_TEST, then you must also update the cached schema password using the administration tool Edit Instance page after you change the password in the database.

The default instance is also used by the Oracle Ultra Search sample query application. You must update the data-sources.xml file, as described in the "Configuring the Middle Tier Component" section of the Oracle Ultra Search User's Guide.

13.2 Creating an Oracle Ultra Search Database User

Oracle recommends that Oracle Ultra Search use its own tablespace to store its own data, as follows:

SQL> create tablespace ultra datafile 'file_location' size x_MB

$ sqlplus system/password@iasdb
SQL> CREATE USER ocs_us IDENTIFIED BY password DEFAULT TABLESPACE ultra
TEMPORARY TABLESPACE temp QUOTA UNLIMITED ON ultra;
User created.
SQL> GRANT resource, connect, wkuser TO ocs_us;
Grant succeeded.
SQL> exit

The WKUSER role gives the user access to Oracle Ultra Search.

13.3 Configuring the Oracle Ultra Search Crawler

This section explains how to configure the Oracle Ultra Search crawler.

Note:

Follow these steps when a new instance is created. In step 10, the paths for Temporary Directory Location and Crawler Logging must be applied to the default instance WK_TEST.

To configure the Oracle Ultra Search crawler:

Create directories on the infrastructure machine where Oracle Ultra Search is installed to hold the temporary and log files.

Go to the Oracle Ultra Search administration tool at the following URL:

http://middletier_host:middletier_port/ultrasearch/admin_sso/index.jsp

When the Single Sign-On page displays, enter the Oracle Internet Directory administrator's username and password and click Login to display the Welcome page.
Click the Create subtab to display the Create Instance page.
Click Create instance.
Enter a New Instance Name and enter the name of the database schema created in "Creating an Oracle Ultra Search Database User" in the Ultra Search Schema Name field.
Click Apply and wait for the page to refresh. A confirmation message displays to confirm the creating of the new instance and the Welcome page displays.
Select the instance you just created from the Select Instance to Manage drop-down list and click Apply to select this instance as your current instance. You receive a confirmation message.
Click the Crawler tab.
Enter the following information:
- Crawling Depth: The maximum number of nested links the crawler will follow. You can limit the depth by selecting the Limit radio button and entering a number in associated field.
- Default Character Set: Choose Standard UTF8 as the default for HTML documents
- Temporary Directory Location: Enter the path to the temporary directory you created in Step 1 of this task
- Crawler Logging: Enter the path to the log directory you created in Step 1 of this task
- Note the value in the Database Connect String field. This value is required when you configure Oracle Files for Oracle Ultra Search in "Setting Oracle Ultra Search Properties".
Click Apply.

13.4 Creating a Web Source

To create a Web source:

Click the Sources tab to display the Create Web Source: Step 1 page.
Click Create Web Source.
Choose a small Web site to index as a test site for Oracle Ultra Search and enter it in the Source Name field.
Click Proceed to Step 2 to display the Create Web Source: Step 2 page.
Click Proceed to Step 3 to accept the defaults and display the Create Web Source: Step 3 page.
In the Starting Address section, enter a URL for a starting address for indexing.
Click Add to add the new address to the Starting Address List.
Click Proceed to Step 4 to display the Create Web Source: Step 4 page.
Click Proceed to Step 5 to accept the defaults and display the Create Web Source: Step 5 page.
Click Proceed to Step 6 to accept the defaults and display the Create Web Source: Step 6 page.
Click Finish to accept the defaults.

The new Web source appears in the Web Source List.

13.5 Creating a Schedule

To create a schedule:

Click the Schedules tab to display the Synchronization Schedules page.
Click Create New Schedule to display the Create Schedule: Step 1 of 3 page.
Enter a schedule name in the Name field.
Click Proceed to Step 2 to display the Create Schedule: Step 2 of 3 page.
Click Proceed to Step 3 to accept the default schedule. This schedules the crawl to execute once a month.
Choose a data source for the schedule to crawl, by doing the following:
1. Select Web from the Get Available Sources for Type drop-down list and click Get Sources.
2. The source that you created in "Creating a Web Source" appears in the Available Sources list.
3. Select the source and click >> to move the source to the Assigned Sources list.
4. Click Finish to return to the Synchronization Schedules page.

The new schedule displays in the Schedules column.

13.6 Scheduling the Oracle Ultra Search Crawler

To schedule the Oracle Ultra Search Crawler:

Click Scheduled in the Status column to display the Synchronization Schedule Status page.
Click Execute Immediately to start the crawler.
In the Crawler Progress Summary and Log Files by Data Source section, click the pencil icon in the Statistics column to display the status and summary for the crawl.

Click the Refresh Status link if you do not see the pencil icon.
To determine if the crawl was successful, verify that Documents Indexed in the Name column has a value other than zero.

13.7 Setting Oracle Ultra Search Properties

After you index documents, you must set the Oracle Ultra Search properties in Oracle Collaboration Suite.

To set the Oracle Ultra Search properties:

Start Oracle Enterprise Manager and navigate to the Oracle Files domain page.
Click Federated Search Configuration to display the Federated Search Configuration page.
Select the Yes radio button for Ultra Search Configured.
Select the Yes radio button for Mail Configured.

Set the Webmail Base URL to

http://middletier_host:middletier_port/um/traffic_cop

Enter the name of the database schema that you created in "Configuring the Oracle Ultra Search Crawler" in the Ultra Search Schema Name field.
Enter and confirm the password that you chose for the database schema in "Configuring the Oracle Ultra Search Crawler" in the Ultra Search Schema Password and Confirm Password fields.
Enter the value that was displayed in step 10 "Configuring the Oracle Ultra Search Crawler" in the Ultra Search Connect String field, prepended with jdbc:oracle:thin:@. For example, jdbc:oracle:thin:@infrastructure_host:1521:iasdb.
Click Apply to receive a confirmation message that the Federated Search configuration has been successfully updated.
After modifying the Federated Search Configuration, you must restart the OC4J_iFS_files instance. You can either restart it from Oracle Enterprise Manager, or execute the following command from the command line:
```
opmnctl restartproc gid=OC4J_iFS_files
```

13.8 Making E-mail Searchable

In order to enable Oracle Ultra Search to search e-mail, you must index e-mail and configure the HouseKeeping process to regularly synchronize the e-mail text index.

This section contains the following topics:

Indexing E-mail
Configuring the HouseKeeping Process

13.8.1 Indexing E-mail

To index e-mail for users as they are created, do so at the domain level, as follows:

Log in to the Webmail client as a user with administrator privileges.
Click the Administration tab.
Click the Domain tab.
Ensure that Domain Settings for Users is selected.
Choose the appropriate Installation and Domain from their respective drop-down lists and click Submit.
Set Mail User Index Type to Indexed for Text and click Submit.

To index e-mail for users that have been created prior to making the above change at the domain level, you must do it on an individual basis, as follows:

Note:

Only new e-mail is indexed. E-mail sent prior to enabling Text Synchronization is not indexed.

Log in to the Webmail client as an e-mail administrator with administrator privileges.
Click the Administration tab.
Click the User tab.
Click Default New User.
Enter the search criteria for the existing user(s) and specify the appropriate e-mail domain from the drop-down list and click Go.
Click the user ID.
Set Mail User Index Type to Indexed for Text and click Modify.

13.8.2 Configuring the HouseKeeping Process

To configure the HouseKeeping process:

Using Oracle Enterprise Manager, navigate to the HouseKeeping page.
Clicking Create or Create Like to create a new Oracle Email housekeeping instance.
Click the new housekeeping instance to bring up the parameters screen and set Text Synchronization to Enabled.
Disable the Pruning and Collection options.
In the Process Sleep Duration field, enter how often the housekeeper should queue messages for indexing, in minutes. For example, if you want the housekeeper to queue messages for indexing every three minutes, enter 3 in the field.
Note:
Set the HouseKeeper to index messages as frequently as clients check for new mail, such as five minutes.
Set Execution Mode to Daemon.
Click Apply.
Return to the HouseKeeping page.
Click Start.

As an ongoing process, housekeeping periodically collects newly arrived messages to Oracle Text for indexing.

13.9 Restarting OC4J_iFS_files

The new Federated Search settings are not used until the OC4J_iFS_files system component is restarted.

To restart the OC4J_iFS_files system component:

On the Federated Search Configuration page in Oracle Enterprise Manager, click the first link in the navigation path under the subtab bar.
Select the radio button next to the OC4J_iFS_files system component.
Click Restart.
Click Yes to confirm that you want to restart.
Click OK to confirm the restart message.

13.10 Testing Oracle Ultra Search

To test Oracle Ultra Search:

Go to the following URL to display the Oracle Files Search window:

http://middletier_host:middletier_port/files/app/FederatedSearch

Log in as the test user to display the General Search page.
In the Item Name field, enter a name for which to search.
In the Content field, enter content for which to search.
In the User Name field, enter the name of the owner of the content to search.
Choose a comparison operator from the Date drop-down list and enter a date string in the associated field against which to search.
Choose Files, Web, or both in the Look In field to indicate which source or sources to search.
After entering the search criteria, click Search.

To search Oracle Files content or Oracle Email, configure them as sources in Oracle Ultra Search.