Join Rules

Join rules determine the linking of a connector space object with an existing metaverse object. Join rules can have two parts: search criteria and resolution. You use the search criteria to compare connector space object attribute values and metaverse object attribute values. The search criteria are listed in the join rules in order of precedence. When you run a management agent, a join search is applied to each object in the connector space, which attempts to find a corresponding object in the metaverse, based on the search criteria.

The following table shows the action taken when a search returns a result.

For this search result	The following occurs
Not one metaverse object is acceptable	The next search criteria in the precedence list is evaluated. If there is no more criteria in the precedence list, processing proceeds to the projection rules, if they exist.
Exactly one metaverse object is acceptable	If there is no rules extension for join resolution, the connector space object is joined, or linked, with the metaverse object. If you have configured a rules extension for join resolution, the rules extension will determine if the object will be joined.
More than one metaverse objects are acceptable	The connector space object is not joined with the metaverse object. If you have configured a rules extension for join resolution, the rules extension will attempt to determine which object to select. Alternately, you can use Preview to diagnose the reason for the multiple search results, and then use Joiner to manually join the connector space object to the metaverse. For more information about configuring a rules extension for join resolution, see the FIM Developer Reference. For more information about Preview, see Using Preview.

The following flow chart shows the sequence in which management agent rules are applied.

Best practices for join rules

Join rules are evaluated in order from top to bottom until a single match is found. When you configure the order of the join rules, start with the most significant (that is, the rule that is most likely to find a unique match) first and work down to the least significant (that is, the rule that is least likely to find a unique match).

To resolve multiple matches, use a rules extension to create a join resolution for each join rule. You configure join resolution in the management agent by using Management Agent Designer. In most cases this is more efficient that moving on to the next join rule. For more information about creating and using rules extensions, see the FIM Developer Reference.

When you define your join criteria, use only attributes that have their Indexed property set to Yes in the metaverse schema. This increases the efficiency of your searches. For more information about indexed attributes, see Using Metaverse Designer.