Published Release Notes

Rollback functionality is limited to latest revision

The rollback functionality in DSM Revision Manager allows customers to roll back the import of the latest revision to their DSM application. Only one rollback is allowed; after a rollback, the rollback option is disabled until the next revision is imported. If importing two (or more) revisions at once then attempting a rollback, the system only rolls back one revision.

Example: a user has a base DSM application. They import Revision 1, and then import Revision 2. Next, they try to roll back both revisions. The system reverts to the Revision 1 version of the application, and the customer cannot revert all the way back to the base application.

Action required by admins for improved search

Starting November 13, 2014, after updating to or upgrading from a prior version to Pega 7.1.7, your system administrator must manually build all indexes for Elasticsearch before the system can automatically switch from Lucene to Elasticsearch.

It is recommended that you begin this manual re-indexing process for all enabled index types as soon as your update or upgrade is complete. The search landing page refers to Elasticsearch settings; during re-indexing, however, Lucene search continues to function to ensure that there is no interruption in search functions. When planning the re-indexing, be aware that the initial re-indexing requires significant resources on the host node and can be a lengthy process, depending on the number of rules and work objects in your system.

See the Pega 7.1.7 Upgrade Guide for information about manually indexing Elasticsearch.

System Management Application displaying listeners that do not require any action

After installing or upgrading to Pega 7.1.7, there is an additional available listener (Data-Decision-DNode-Service:Default) that does not require any action. This listener can be safely ignored and you should not use the System Management Application to manage any of its operations or state.

• The Data-Decision-DNode-Service:Default listener is always running on every PRPC node.
• The state of this listener is internally managed by PRPC. Starting or stopping it through SMA does not have any effect on its state.

Open ports required for cluster communication

In a Pega 7 system, there can be multiple servers, or nodes, and each node can contain multiple JVMs.  In Pega 7.1.7, the port range 5701-5800 needs to be left open for cluster communication.   The number of available ports in this range needs to be greater than or equal to the greatest number of JVMs on any one node in the cluster.  So for example, if there are 3 JVMs on one node, and 7 JVMs on another node, there need to be at least 7 ports available.  By default, the system will begin with port 5701, and then look for the next port in the sequence (5702, followed by 5703).

If these ports are not available, you will see the following error:

ERROR   - PegaRULES initialization failed. Server: ABCDEF123

com.pega.pegarules.pub.context.InitializationFailedError: PRNodeImpl init failed

< . . . Java dump here . . . >

Caused by: com.hazelcast.core.HazelcastException: ServerSocket bind has failed. Hazelcast cannot start! config-port: <port>, latest-port: <port>

Using Kerberos authentication with your database

Pega 7 supports Kerberos functionality. Kerberos is a computer network authentication protocol which allows nodes communicating over a non-secure network to prove their identity to one another in a secure manner. 

To use Kerberos for authentication, you must use the command line to install or upgrade Pega 7.

To use Kerberos authentication:

1.  Change the setupDatabase.properties file.

a.  In the “Uncomment this property section” of the file, uncomment the jdbc.custom.connection.properties property.  Based on your security infrastructure, different properties may be required as parameters to this property; provide the needed properties as semicolon-delimited name/value pairs:
 

jdbc.custom.connection.properties=prop1=val1;prop2=val2;prop3=val3;
 

Example:  For an installation on a MSSQL database server from a Windows client machine (where both machines belong to the same Windows domain), using the Microsoft JDBC driver, the property may be set as follows:
 

jdbc.custom.connection.properties=integratedSecurity=true;

b.  Comment out all the username and password properties where they occur in the jdbc.custom.connection.properties file, so that they appear as follows:

# pega.jdbc.username db username
# pega.jdbc.password db password

[lines removed here]

# pega.jdbc.username=ADMIN
# pega.jdbc.password=ADMIN

2.  Set up your database to enable Kerberos functionality.  This may include additional vendor-specific JDBC driver configuration, or other setup procedures.  Check the documentation from your database vendor to determine what Kerberos setup is needed for your database.

3.  Run the command line installation or upgrade by following the instructions found in the Pega 7 Deployment guides.

Run cleanup.bat/sh script only before upgrading

Prior to upgrading the rulebase, you can optionally run the cleanup.bat/sh script to remove older rules from the database.

Run this script before you upgrade your rulebase, or the script may delete needed rulesets. For more information about running the cleanup.bat/sh script, refer to the Upgrade Guide specific to your release version.

Recommended heap sizes

The heap is a storage area in the Java virtual machine (JVM) allocated for both short-term and long-term (shared) object storage. If the server does not have enough memory allocated to run Pega 7, the system can hang without an error message. If this occurs, your values need to be higher than the recommendations based on your server hardware and the number of other applications on the server.

Pegasystems recommends using these settings:

• Initial Heap Size (Xms): 1 GB
• Maximum Heap Size (Xmx): 12 GB (8 GB minimum)

If your application server is using the Oracle JVM, also add the PermSize and MaxPermSize settings:

• PermSize (-XX:PermSize): at least 256MB
• MaxPermSize (-XX:MaxPermSize): at least 512MB

Set the JVM memory options to increase the amount of system memory allocated to the application server running Pega 7 by selecting Designer Studio > System > Operations > System Management Application and then selecting a listed node.

See the Pega 7.1.7 Installation Guides for more information about how to make these setting changes.

When using Oracle 12.1.0.1, left outer joins in reports may return incorrect results

When using  Pega 7 with Oracle 12.1.0.1 , reports that use left outer joins may return incorrect results.

This is an Oracle known issue.  To resolve this behavior, upgrade to Oracle 12.1.0.2, apply the Oracle patch 16726638 by requesting it directly from Oracle, or, for Microsoft Windows installations, apply Windows Bundle patch 12.1.0.1.15 or later.

For more information in this behavior, see: https://support.oracle.com/epmos/faces/DocumentDisplay?_afrLoop=392085808201943&id=1957943.1

Custom database (DB) triggers are dropped during upgrade

The latest version of Pega 7 improves performance by no longer using database triggers to assist with System Pulse and Data-Rule-Summary processing, which is now done within the Pega 7 engine. As a result, DB triggers are no longer installed on either the pr_sys_updatescache or pr4_rule_vw tables.

When upgrading to the latest version of Pega 7, if you had previously implemented a custom database (DB) trigger on these tables, or a custom DB trigger that refers to these tables, it is removed during the upgrade process. No custom triggers are removed unless they reference these tables.

If you have custom DB triggers that reference the pr_sys_updatescache or pr4_rule_vw tables and perform other processing, those triggers must be reimplemented. When doing so, you must be careful to not modify the pr_sys_updatescache or pr4_rule_vw tables.

For more information, see Startup check removes custom DB triggers.

Convert data instances into decision parameter rules

The Convert groups wizard is available in the Hierarchy landing page when the list of your propositions include decision data records.

Click Convert groups at the top of the landing page to use the wizard and to convert data instances into decision parameter rules. This conversion is necessary to make the existing proposition data instances available for revision management.