SR-A91805 · Issue 265722
Archive/ Purge JVM dependence removed
Resolved in Pega Version 7.2.2
If the JVM was shut down in the middle of the purge and archive process, some SLA queue items remained in a 'now-processing state' and were not automatically picked up by the SLA agent for rescheduling when the JVM was brought up again. This has been fixed by redesigning the the Archive/ Purge flow to run in the background in queue batches so SLA queue items will be updated immediately with their next goal. With this approach, updates of SLA queue items will not depend on purge and archive process duration, and SLA queue items will always be in a scheduled state.
SR-A97582 · Issue 270053
Resolved error in importing code jar with tracer on
Resolved in Pega Version 7.2.2
Importing a code jar with tracer on intermittently failed and showed a blank screen. This was caused by pxCheckCodeSet closing the archive even through import is in progress, and has been fixed by modifying the CheckCodeSet activity to mark the import progress as true so that passivation doesn't cause archive to be closed.
SR-B469 · Issue 273774
Improved performance for installing schema related raps
Resolved in Pega Version 7.2.2
Performance improvements have been made relating to installing schema related raps.
SR-A99908 · Issue 270284
Purge export logic updated to handle TimeCreated keys that differ by milliseconds
Resolved in Pega Version 7.2.2
Purge archive was failing during export of the archives. Analysis showed that when exporting keys of history class mapped to a table with no blob and pxTimeCreated as Key column, mapping this to a column of Date type on Oracle caused the pzinskeys of history- instances to be recalculated and the millisecond part of the time stamp was rounded in the key to 000.This caused keys which just differed in milliseconds to become duplicates and export failed. This has been resolved by modifying the export logic to handle keys which differ in milliseconds for Blobless History class instances.
SR-A92640 · Issue 266679
Upgrade rules considers DADT instances for mapping
Resolved in Pega Version 7.2.2
When migrating and upgrading the rules tables to a new schema, the _upgradeRuleBase task was trying to remap some work classes/tables to pr_other. This was due to the class name being compared with case sensitivity when the site had DADT instances for which the class name was not same as the class instance. To resolve this, the check has been made case insensitive so DADT instances are not missed.
SR-A102766 · Issue 269648
db2zOS tablespace script logic updated
Resolved in Pega Version 7.2.2
When running the generateddl script against a db2zos database, the tool determines for each table if the current tablespace page size needs to be increased before the DDL changes can be applied. For the pc_work_social table, while the tool included the size of new columns when it calculated the projected table size, it did not take into account the altering of existing columns. This has been fixed by tightening up the JDBC resource handling.
SR-A98103 · Issue 266253
Scheduled recurring Purge/Archive set to reuse specified start time
Resolved in Pega Version 7.2.2
When scheduled to run every day at a particular time, the Purge/Archive function started later each time. This was due to the function using the Agent running time for the rescheduling, starting the next day at the time it ended. The function has now been modified to use the customer-specified start time instead of calculating the Agent time.
SR-A86384 · Issue 258433
Diagnostic logging added for "No stream to display" error
Resolved in Pega Version 7.2.2
Diagnostic logging has been added to better address the "No stream to display" error on the UI. When enabled, the logger will capture the postDataBytes and corresponding postDataString that are returned.
SR-A21743 · Issue 256234
Support added for WebLogic 12.2.1
Resolved in Pega Version 7.2.2
Support has been added for WebLogic 12.2.1, which introduces a new Class Loader between the application and System Class Loader.
SR-A76628 · Issue 255871
Forced logging type changed to avoid incorrect alerts from WebLogic
Resolved in Pega Version 7.2.2
When using PRPC with WebLogic, server restarts were generating the notice that Emergency messages were present in the Server console log file. This was due to WebLogic treating logs with level greater than 1000 as emergency while PRPC was using level greater than 1000 for forced logs (infoForced and warnForced) to ensure that forced logs were not skipped in any log level setting except for level OFF. As WebLogic does not have any equivalent for forced logging, it interpreted this as an emergency. This behavioral conflict has been resolved by changing the PRPC logs from infoForced to info.