Support for Hotfix Manager in service-enabled system management
Valid from Pega Version 7.2.1
You can manage hotfixes by using a REST service from Hotfix Manager in the service-enabled system management utility. With service-enabled system management, you can manage hotfixes simultaneously on more than one Pega 7 Platform system. You can manage the following hotfix operations: install, scan, commit, rollback, and GenerateDDL.
For more information, see Managing hotfixes (service-enabled).
Improved Pega 7 Platform upgrade and update time
Valid from Pega Version 7.2.1
The Pega 7 Platform upgrade and update tools have been improved to streamline indexing. Upgrading and updating are faster than with previous releases.
Option to resume failed upgrades and updates
Valid from Pega Version 7.2.1
When you upgrade or update to Pega 7.2.1, you can opt to have the system resume the deployment after a failure to avoid restarting the deployment and to save time. The deployment resumes at the point of failure and does not repeat any successful steps. This functionality is not supported for new installations. For more details, see the Pega 7 Platform Upgrade Guide and the Pega 7 Platform Update Guide.
Open access to quiesced servers when using immediate drain
Valid from Pega Version 7.2.1
When using the immediate drain method to perform quiesce on a node, any operator can now access a quiesced server for root cause analysis or remediation, regardless of their user role or privileges. For quiesced server access when using the slow drain method, you must still include either the PegaRULES:HighAvailabilityAdministrator or PegaRULES:HighAvailabilityQuiesceInvestigator user role in a user's access group. Administrators using the slow drain method without either of these user roles are exiled from the quiesced node and are redirected to an active node.
For more information, see Immediate drain available for quiesce when using high availability.
New passivation and activation activities
Valid from Pega Version 7.2.1
Two new activities have been implemented, pyPrePassivation and pyPostActivation, which are run every time operator passivation occurs, regardless of the node quiesce state. These activities save and restore the requestor state during passivation and activation. In order for the pyPrePassivation or pyPostActivation activity to be run as expected, you must implement the activity by using the correct Applies To class (@baseclass).
The previous activities that were run during the quiesce process, pyPreQuiescePassivation and pyPostQuiesceActivation, have been deprecated.
For more information, see the Pega 7.2.1 High Availability Administration Guide.
Support for reserved keywords in database column names
Valid from Pega Version 7.2.1
Databases use reserved keywords such as Group, Order, Rank, and Date to define and manipulate data. The list of reserved keywords is specific to each database. Column names can now include reserved keywords.
Define adaptive model context
Valid from Pega Version 7.2.1
You can define model identifiers in the Adaptive Model rule instance, which serve as partitioning keys, by selecting properties from the top-level SR class of your application. The Adaptive Decision Manager (ADM) server uses model identifiers to create adaptive models that use the same configuration, but collect different adaptive statistics. Use model identifiers in the Adaptive Model rule to create adaptive models for your purposes.
For more information, see the Context tab on the Adaptive Model form.
High number of busy waits when using Oracle SecureFiles
Valid from Pega Version 7.2.1
You might experience a high number of buffer busy waits when using Oracle SecureFiles for BLOB storage. An Oracle buffer busy wait occurs when a session cannot access a database block in the buffer cache, because another session is either reading the block into the buffer or changing the buffer. As a best practice, use Oracle 11.2.0.4 or later, because earlier versions have more problems with SecureFiles. In addition, to minimize the number of busy waits, set the hidden Oracle parameter _securefiles_concurrency_estimate to 50, and partition the table by using a composite range hash partition that is based on the pxCreateDateTime and pzInskey columns.
For more information, see Troubleshooting: High number of busy waits when using Oracle SecureFiles.
DSM rulesets become part of the PegaRULES application
Valid from Pega Version 7.2.1
The Pega-DecisionArchitect, Pega-DecisionEngine, Pega-BigData, and Pega-NLP rulesets were moved from the PegaDM application record to the PegaRULES application record. This change makes decision management functionality available for Pega 7 Platform users by default without any additional configuration required.
For more information, see DSM rulesets become part of the PegaRULES application.
HDFS data sets support Parquet files
Valid from Pega Version 7.2.1
The Parquet file format has been added to the list of supported file formats for data storage in the Hadoop Distributed File System (HDFS) data set record. On the Pega 7 Platform, you can now access an HDFS data set that stores data in the Parquet file format and perform various operations on that data set.
For more information, see Big Data enhancements in decision management.