Published Release Notes

Ability to train the text analytics model based on email attachments

You can now train the text analytics model for Pega Email Bot™ based on training data from email attachments. As a result, you improve the natural language processing (NLP) analysis for your email bot so that your system detects the correct language, sentiment, topics, and entities from both the body of emails and their attachments. By default, you train the text analytics model based only on training data from the body of emails.

For more information, see Enabling training the model based on email attachments.

Upgrading to the secure threading mechanism for email bots

In Pega Platform™ version 8.6, Pega Email Bot™ includes a more secure threading mechanism to help track emails from customers and other stakeholders in separate threads for an email triage case.

Upgrade impact

If you upgrade from Pega Platform version 8.5 or earlier, in which you configured an Email channel, perform the following steps to ensure that your system uses the new secure threading mechanism:

• Update the service email rule that the system uses to send an email reply as the initial acknowledgment.
• Update the email reply template in the data transform rule that the system uses when a customer service representative (CSR) sends the reply.

For more information about creating an initial acknowledgment email and email reply template, see Creating outbound email templates. For more information about the secure threading mechanism, see Use a secure threading mechanism in emails.

What steps are required to update the application to be compatible with this change?

For the initial acknowledgment email used by your email bot, update the service method for your email listener rule. On the Response tab for this service email rule, expand the Message contents section. In the Message data section, you specify the rule that defines the structure of the content of the email body. In Pega Platform version 8.6, you use for this purpose the pyEmailAcknowledgement correspondence rule that takes into account the selected built-in template. This template includes the security code tag that the system uses for the secure threading mechanism. If your application uses a different rule in the Message data section, update this definition to match one of the built-in correspondence template rules, for example, EmailAckTemplate_Clear.

The pySetEmailBotReplyTemplate data transform rule sets the name of the email correspondence rule that the system uses as the email reply template. If you do not want to use the default approach using the Classic, Cobalt, or Clear outbound email template themes, override this data transform rule to set the email correspondence rule name for the Param.ReplyTemplate target in the Source column field.

For more information about how to update the service email rule and the data transform rule to ensure that your system uses the secure threading mechanism, see Upgrading to the threading mechanism available in the 8.6 version.

Improved indexing performance by gradual retrieval of data

Search functionality in Pega Platform™ versions from 8.4.5 to 8.6 now includes the option to improve indexing performance when reading query results from a large table in a database. For example, to load the recommended 50 records at a time, in the Pega-SearchEngine ruleset, create the indexing/distributed/fetchsize dynamic system setting, and set the value to 50.

Creating the fetchsize dynamic system setting ensures that your system does not crash while indexing classes with numerous instances.

For more information, see Creating a dynamic system setting.

Data model inheritance does not depend on ruleset context

Search and Reporting Service (SRS) in Pega Platform™ 8.6 now provides an improved method of synchronizing data model by distinct users. In applications built on earlier versions of Pega Platform, some users might encounter difficulties in indexing data. Now, the data model that the system sends to SRS does not depend on access privileges. With this enhancement, each user of your application is eligible to synchronize data.

Tomcat log files moved to work directory

For Tomcat application server users, the default path for PRPC logs has been changed to the "work" directory for Catalina. For example:

<tomcat>\work\Catalina\localhost\prweb

This directory can be cleaned up during administration. To avoid loss of files, you can redirect logging to a different directory after installation:

1. Navigate to the <tomcat-home>\webapps\prweb\WEB-INF\classes directory.
2. Open the prlogging.xml file.
3. Locate the "appender" definitions section that contains the "@{web.tmpdir}" string
4. Update the FileNamePattern values to point to the new directory.

For more details, refer to the “Change the default path to PRPC logs” section in the Tomcat installation guide for your database.

Supported JDBC drivers

Pega 7.1.1 supports Java 6 and Java 7 SDKs only. Ensure that the JDBC driver you are using supports those versions of Java. See your database installation or upgrade documentation for additional information.

* JDBC drivers are only available from the DB2 for z/OS product. Additionally, you need db2jcc_license_cisuz.jar in the same directory as db2jcc4.jar.

** The db2jcc.jar has been deprecated and may not be supported in future releases.

*** This .jar file is required for PostgreSQL version 9.1.x.

Additional upgrade scripts for new columns

In Version 6.2, additional columns (Major, Minor, and Patch) containing ruleset Version information were added to 15 tables in the PRPC database. When upgrading from a pre-Pega 7 release, scripts must be manually run after the upgrade to populate these columns.

These scripts are located in the Resource Kit, under Additional Upgrade Scripts.

• If upgrading to a single schema, run both scripts.
• If upgrading to a split schema, run the _data script against the data schema and run the _rules script against the rules schema.

Choose the scripts for your database type:

• db2zos_rulesetversion_columns_data.sql
• db2zos_rulesetversion_columns_rules.sql

• mssql_rulesetversion_columns_data.sql
• mssql_rulesetversion_columns_rules.sql

• oracle_rulesetversion_columns_data.sql
• oracle_rulesetversion_columns_rules.sql

• postgres_rulesetversion_columns_data.sql
• postgres_rulesetversion_columns_rules.sql

• udb_rulesetversion_columns_data.sql
• udb_rulesetversion_columns_rules.sql

Required Oracle optimization parameter

To optimize performance, set the Oracle parameter optimizer_index_cost_adj to a value between 20 and 25. If this value is not set, the system can run exceedingly slowly, potentially blocking users from login.

Slow WebSphere performance on Mac platforms

Use the following JVM setting to improve WebSphere performance on Mac platforms:

-Djava.net.preferIPv4Stack=true

Turn off saved connections in WebLogic

PRPC caches connections in thread-local storage. When using a WebLogic application server, set database/savedconnections to FALSE to avoid incorrectly saved, container-managed transactions. You can make this update through JNDI settings or by adding an entry to the prconfig.xml, for example:

<env name="database/savedconnections" value="false" />