SLA Endpoint Lookup mediation primitive

Use the SLA Endpoint Lookup mediation primitive to look up endpoints referenced from an active service level agreement (SLA) between a service consumer and a service provider.

Introduction

If you use IBM Process Server to integrate your enterprise and model your business in WSRR, using its Governance Enablement Profile (GEP), you can select dynamic endpoints based on a number of factors modelled in that profile. For example:

Whether the consumer of the endpoint has a valid SLA for the endpoint
Whether the particular SLA is active
Whether the endpoint is online
Whether the endpoint has a certain required classification; for example, whether it is Production or Development.

The SLA Endpoint Lookup mediation primitive has one input terminal (in), one fail terminal (fail), and two output terminals (out and noMatch). The in terminal is wired to accept a message and the other terminals are wired to propagate a message. If the information passed on the incoming message successfully finds a matching endpoint in WSRR, the out terminal is fired. However, if a matching endpoint is not found, the noMatch terminal is fired.

Usage

To find a matching endpoint in WSRR, the SLA Endpoint Lookup mediation primitive uses information in the incoming message. The endpoint is matched on a number of parameters, the first three of which are system defined:

Consumer Identifier. This field can be a literal value, or can be passed as part of the message, and identifies the consumer of the target endpoint. The consumer identifier is a field in the CapabilityVersion, ProcessVersion, ApplicationVersion or ServiceVersion representing the consumer of the target service. For more information, see the WSRR documentation on the Governance Enablement Profile.
Context Identifier. This field can be a literal value or can be passed as part of the message and identifies the context under which the consumer's invocation of the target endpoint occurs. The context identifier is a field on the SLA representing the agreement between the consumer and provider of the target service. For more information, see the WSRR documentation on the Governance Enablement Profile.
Endpoint Classification. This field can be a literal value, or can be passed as part of the message, and indicates a refinement in the selection of the target endpoint. All endpoints in WSRR can be classified. For example, they can be Test endpoints or Production endpoints. This classification enables a more fine-grained search. An example of a Production endpoint is http://www.ibm.com/xmlns/prod/serviceregistry/6/1/GovernanceProfileTaxonomy#Production.

Add extra parameters into the search to add further refinement of the endpoint selection. Use the table of user defined parameters under the Advanced tab for this purpose to supply extra selection parameters. However, if you add extra parameters, the named query installed on the WSRR system must be altered to match these additions.

Note: If you are using a WSRR Version 8.0 governance enablement profile, this mediation primitive behaves differently to that in previous versions. The SLA Endpoint Lookup mediation primitive uses a WSRR named query called SLAEndpointLookup. In a query that uses the WSRR Version 8.0 profile only service level definitions that are in the subscribable state are considered. The new behavior ensures that the WSRR objects that are being considered have reached a suitable level of approval. If you want to reproduce the behavior of previous versions, where this state is ignored, you can choose either of the following options:

Continue to use the same governance enablement profile you used on a previous version of WSRR.

Modify the named query on the supplied WSRR Version 8.0 profile by completing the following steps:

Log onto the WSRR by using the web user interface.
Change to the Configuration perspective.
Go to Active Profile > Named Queries.
Select SLAEndpointLookup.

Replace the <xpath> query with the following code:

<xpath>/WSRR/GenericObject[classifiedByAnyOf(.,
  'http://www.ibm.com/xmlns/prod/serviceregistry/profile/v6r3/GovernanceEnablementModel#CapabilityVersion') 
   and gep63_consumes(.) and @gep63_consumerIdentifier='%1']/      gep63_consumes(.)[classifiedByAnyOf(.,
'http://www.ibm.com/xmlns/prod/serviceregistry/lifecycle/v6r3/LifecycleDefinition#SLAActive') 
   and @gep63_contextIdentifier='%2']/gep63_agreedEndpoints(.)/ gep63_availableEndpoints(.)[classifiedByAnyOf(.,
'http://www.ibm.com/xmlns/prod/serviceregistry/lifecycle/v6r3/LifecycleDefinition#Online') and classifiedByAnyOf(.,'%3')]
</xpath>

Select OK.

The new named query is now active.