Published Release Notes

Primary key constraint error when using JBoss

A primary key constraint violation exception might occur in the following situation: you use JBoss EAP 6.4 and SQL Server 2012 with a JDBC 4 driver, and you use two or more file listeners or JMS MDB listeners with concurrent threads that create work objects that use multiple files. If a primary key constraint violation exception occurs in this situation, update the data source configuration file to set the jta parameter to true. For example, set <datasource jta="true" jndi-name-"java:/jdbc/PegaRULES>.

Decision Analytics portal replaces Predictive Analytics Director portal

The Decision Analytics portal replaces the Predictive Analytics Director (PAD) portal, the Text Analytics landing page, and the Adaptive Models Reporting landing page. In this new work area for predictive analytics and text analytics, the business scientist can control the full model life cycle by building predictive models, importing PMML models, building sentiment analysis and text classification models, creating and monitoring adaptive models, and updating any existing models.

For more information, see Overview of the Decision Analytics work area.

Node classification

By using the node classification feature, which segregates nodes by their purpose, you can now configure one node with multiple node types. When you start a node with multiple node types, the node runs a set of agents and listeners that are associated with the specified node types. From the Node Classification landing page, you can view and configure the agents associated with node types, view listeners associated with node types, and open Agent rule forms, agent schedules, and Listener rule forms.

For more information, see Node Classification, Configuring node types for agents and listener and Node classification.

Support for Couchbase Server NoSQL database

Pega^® Platform can now read data from and write data to a Couchbase Server database. Couchbase is a NoSQL database that you can use when a NoSQL database better meets your scalability and processing requirements.

For a list of supported and unsupported features, see Couchbase Server support.

Node types renamed after upgrade from Pega 7.2.2 to Pega Platform 7.3

If you used node classification in Pega^® 7.2.2, when you upgrade to Pega 7.3, node type names are automatically changed to a new name when you start a node with a node type.

• UniversalNode becomes Universal
• WebUserNode becomes WebUser
• BackgroundProcessingNode becomes BackgroundProcessing
• BIXNode becomes BIX
• SearchNode becomes Search

If you created agents in Pega 7.2.2, agent schedules are also automatically updated with a new name during the upgrade. For more information, see Customized agent schedules for standard Pega Platform agents must be updated after Pega 7.2.2 to Pega Platform 7.3 upgrade and the appropriate Deployment Guide.

Event type browsing on the Customer Movie landing page

Event Browser is a new option on the Customer Movie landing page that provides insight into the history of customer interactions. You can retrieve data from the Event Store data set for a particular customer or a group of customers, display the events in a timeline, and browse a list of event types.

For more information, see Browsing events for a customer or a group of customers.

SystemCleaner agent is split into node-level and cluster-level agents

The Pega-RULES:SystemCleaner agent has been split into node-level and cluster-level agents to avoid database performance issues. The new node-level agent, Pega-RulesEngine:NodeCleaner, runs on all the nodes in the cluster, while the new cluster-level agent, Pega-RulesEngine:ClusterAndDBCleaner, runs on only one node in the cluster. If you customized the agent schedules of the Pega-RULES:SystemCleaner agent, you must manually update the agent schedules of the new agents because customized schedules are not retained.

For more information, see Agent schedule data instances – Completing the Schedule tab.

Select multiple nodes for the Adaptive Decision Manager service

You can select multiple Pega Platform nodes and assign them to the Adaptive Decision Manager (ADM) service as worker nodes. Adaptive learning is no longer restricted to a single server, which provides resilience and scalability. You can add or remove nodes from the ADM service while the system processes customer responses.

For more information, see Changes to Adaptive Decision Manager.

Greater cluster stability with client-server clustering topology

You can now deploy Pega^® Platform in the new client-server clustering topology that uses Apache Ignite, which increases stability in large clusters and provides independent scalability for both servers and clients. This technology separates the client nodes, which are responsible for Pega Platform processes, from the server nodes, which are responsible for cluster communication and distributed features. If your production environment consists of more than five nodes in the cluster, it is recommended that you switch to the new client-server topology.

For more information, see Apache Ignite client-server clustering topology and the appropriate Deployment Guide.

Performance enhancements to the Visual Business Director data set

When you create instances of the Visual Business Director data set, you can specify the granularity of data aggregations and select dimension levels to reduce the Visual Business Director memory footprint and query time. In addition, the Actuals Visual Business Director data set on the Data-pxStrategyResult class has been replaced with a regular Decision Data Store instance. You can now quickly update or synchronize records in the Actuals data set with the Interaction History data set.

For more information, see Data Set rule form - Completing Data Sets and Aggregation on the Visual Business Director data set.