Child pages
  • JDBCDataConnector

The Shibboleth 1.x software has reached its End of Life and is no longer supported. This documentation is available for historical purposes only.

Skip to end of metadata
Go to start of metadata

Configuring a Relational Database Connector

The relational database connector allows you to pull attributes from data stores that can be access through a Java JDBC driver (which is nearly all relational databases). This connector pools JDBC connections in order to enchance performance. See the advanced configuration section in order to disable this.

Data Connector Basics

All data connectors are configured in the IdP's resolver.xml configuration file.

Each connector is defined with an XML element that requires an id attribute. This attribute is used to reference the connector from other connectors and attribute definitions. To make future maintenance easier we encourage you to use an meaningful name for id attribute.

Configuring the Connector

  1. Create a JDBCDataConnector element, provide it an id, and the follow attributes:
    • dbURL - the JDBC url for your database including the userid and password used to connect to the database
    • dbDriver - the full qualifed class name of the JDBC driver to use (must be on your classpath)
    • useCallableStatement - true if you're using SQL stored procedures (known as callable statments in Java), this attribute is optional
  2. Create a Query child element whose content is the SQL query used to fetch the attributes. You may use a single question mark in place of a parameter value in order to indicate that the principal's name should be used.

Also, be sure that you have the JDBC driver for your database accesible on the IdP's classpath. One approach for doing this is to place the JDBC driver jar in Tomcat's common/lib directory.

Example Configuration

<JDBCDataConnector id="studentSystem"
	 <Query>select entitlement from foo where name = ?</Query>

Advanced Configurations

The JDBCDataConnector element also supports the following attributes:

  • maxActive - maximum number of connection that can be in use at one time or a negative number indicates there is no limit
  • maxIdle - the max number of idle connections that may be open in the connection pool or a negative number indicates there is no limit
  • maxWait - maximum number of seconds to wait if all the connections are currently in use before failing or a negative number indicates there is no limit
  • retryInterval - number of seconds, after a connector has failed (was unable to make any connection to the database), it should wait before retrying
  • minResultSet - minimum number of results that may be returned from a query or an exception is raised
  • maxResultSet - maximum number of results that may be returned from a query or an exception is raised
  • validationQuery - a query that may be used to validate that a connection is still active

At times it is necessary to use attributes pulled from other connectors with the query for this connector. The statement creator allows you to do this. To set it up do the following:

  1. In the Query element use question marks to indicate where attributes values are needed
  2. Create a StatementCreator element, as a child of JDBCDataConnector, with a class attribute of edu.internet2.middleware.shibboleth.aa.attrresolv.provider.DependencyStatementCreator
  3. For each value in the query create a Parameter element, as a child of StatementCreator with the following attributes:
    • type - The SQL object type of the attribute to be used, acceptable values are: String, Integer, Byte, Double, Float, Long, Short, Boolean, Date, Blob, Clob
    • attribtueName - The name of the attribute whose value will be used
    • nullMissing - Indicates whether the value null should be used if the attribute value is null, acceptable values are: true, false
    • connectorID - The ID of the connector the attribute value may be retrieved from

The ordering of the Parameter elements must equate to the order of parameters in the query.

Example Configuration

<JDBCDataConnector id="db2"
	 <DataConnectorDependency requires="echo"/>
	 <AttributeDependency requires="urn:mace:dir:attribute-def:eduPersonEntitlement"/>
	 <Query>select date from foo where principalName = ? and entitlement = ?</Query>
	 <StatementCreator class="edu.internet2.middleware.shibboleth.aa.attrresolv.provider.DependencyStatementCreator">
		  <Parameter type="String" attributeName="eduPersonPrincipalName" connectorId="echo" nullMissing="false" />
		  <Parameter type="String" attributeName="urn:mace:dir:attribute-def:eduPersonEntitlement" nullMissing="false" />

Data Connector Dependencies

In order to use attributes from other connectors or attribute definitions you need to make sure those dependencies are resolved before this connector is called.

If your connector depends on another connector create a child element, of main connector element, called DataConnectorDependency and give it an attribute of requires whose value is the id of the connector this connector depends on.

If your connector depends on an attribute definition create a child element, of the main connector element, called AttributeDependency and give it an attribute of requires whose value is the id of the definition this connector depends on.

Failover Dependency

A connector can specify a special type of dependency, called a failover dependency. If the data connector is unable to fetch any attributes the connector specified in the failover dependency will be invoked.

To specifiy a failover dependency create a child element, of main connector element, called FailoverDependency and give it an attribute of requires whose value is the id of the connector this connector depends on.

Error Propagation

Not every data connector may, or need, work for every user in the system. However, the connector may emmit an error if it is unable to find any entries for the user or if it depends on another connector or attribute definition that didn't apply to the current user.

To suppress these errors, so that they don't stop the attribute lookup process, add the follow attribute and value to the main connector element:

  • propagateErrors - with a value of false

Data Caching

Data connectors cache its information for the length of one attribute request; until all connectors and attribute definitions have been evaluated and their information retrieved. You may optionally have the connectors cache their data for a fixed period of time. This can increase performance but will result in a lead time for frequently changing data.

To enable this longer-lived cache add the following attribute to the main connector element:

  • cacheTime - the length of time, in seconds, to cache the attributes fetched by this connector
  • No labels