Create the MSSQL data source for storing content
You must create an MSSQL data source to store knowledge content that is displayed on the portal, such as white papers, press releases, and FAQs. If you plan to use multiple portals, you can create one data source for all of them, in which case the same content can be exposed on all portals, or you can create a separate data source for each portal.
- In Microsoft SQL Server, create a database for knowledge content. For example, you could create the CustomerKnowledgeContent or PortalKnowledgeContent database.
- Back in the Farm Manager, on the Data Sources tab, click New.
-
Specify this information:
- Name
- Specify a data source name; for example, CustKnowledgeContent.
- Type
- Specify MSSQL.
- Click OK. The data source record is added in the tree.
- Click Edit.
-
Specify this information:
- Description
- Optionally, you can provide a description for this data source.
- Server Instance
- Specify the name of the Microsoft SQL Server instance to use when connecting to the database server. For a single-server test or development environment, this can be the machine name.
- Integrated Security
- This option is available for only installations with a Windows
or Windows domain user as the database access user account.
To allow IPF to access SQL Server without specifying a user ID and password, select this option. This allows IPF to use Windows authentication.
If you are using the SQL Server FILESTREAM feature to facilitate large BLOBs, then you must select this option. Also, the IIS web application must have permissions to access the location of the BLOB. See http://technet.microsoft.com/en-us/library/bb933993(v=sql.105).aspx. Also see “BLOB Handling for MSSQL” in the portal customization guide.
- User ID
- For installations with a Windows or Windows domain user for
the database access user account, if integrated security is not
being used, you must specify a SQL Server authentication user ID for
connecting to the SQL Server database. This must be a user with dbo
permissions in the database.
Also for installations with a control DB user for the database access user account, you must specify a SQL Server authentication user ID for connecting to the SQL Server database.
- Password
- Specify the password for the user ID above.
- Database
-
- Select the database created in step 1.
- Click Save.
- Click Manage Application Schema.
- Click Enable schema management.
- In the Originator field, select the current back
office originator, for example SL 9.01.x, and in the Version
field, select the current version. Note: If the current originator is not available, you may have installed a patch version. See Importing the portal content patch build.
- In the Data Store Type section, click Add.
- In the drop-down field, select SLPortalSchema.
- Click Save.
- Click Edit.
-
Specify this information:
- Default SQL Schema
- Optionally, specify the schema to use when accessing objects such as stored procedures, tables and views. SQL performs slightly better when the schema is specified. When paired with a Mongoose-based database, this option is disabled and "dbo" is always used.
- Query Timeout
- Specify the number of seconds the portal should wait for a request to be serviced by the Server Instance before giving up and returning an error. If set to zero, the portal will wait forever.
- Pooling-Related Options
- The MSSQL data source uses ADO.Net to connect to Microsoft SQL
Server. ADO.Net supports a connection pooling feature to increase
performance. For more information, see: http://msdn.microsoft.com/en-us/library/8xx3tyca(v=vs.90).aspx.
Unless your servers are operating under high-load conditions, you can usually leave the defaults for these options unchanged. These configuration options are available:
- Min Pool Size: Specify the minimum pool size. The default value is zero.
- Max Pool Size: Specify the maximum pool size. The default value is 100.
- Minimize Pools: When this check box is selected, this data source attempts to avoid pool fragmentation (caused by many databases) by implementing the procedure described in "Pool Fragmentation Due to Many Databases" on the web page cited above. It connects to the master database on the server first and then switches to the configured data base.
- Paired Back Office Data Source
- Optionally, you can configure this data source to be used as a stand-in for the specified Mongoose-based data source in certain scenarios. See “Pairing” in the portal customization guide.
- Use Default Values as Initial Values
- This setting is optional. When this check box is selected, new records created in collections associated with this data source have their properties pre-populated with default values as configured for the table in the database. The values used are taken from the COLUMN_DEFAULT column of the INFORMATION_SCHEMA.columns view. Only column defaults that can be converted to the data type of the column are used. If the column default includes a SQL expression, the default is considered null and the application is responsible for setting the initial value.
- Click Save.
- If you plan to create a separate knowledge content data source for multiple portals, repeat this process for each. Give each database and data source a unique name, such as VendKnowledgeContent and ReselKnowledgeContent.