SR-B36632 · Issue 294145
Improved logic for cloned operators and workpools
Resolved in Pega Version 7.3
The workpools available in a new application were not updating correctly due to a missed use case related to multiple cloned operators and the respective multiple cloned workpools being added to the built-on operator. To better handle this scenario, the following changes have been implemented: 1) If the workpool originally listed on the access group to be cloned derives from a class in the list of classes to be created, then the entry will be updated with the new work pool name. 2) If the workpool originally listed on the access group to be cloned does NOT derive from a class in the list of classes to be created, then the entry will be removed from the new access group. (for example, if the user selected no case types and the old access group had multiple work pools listed, then none of those workpools will be present on the new access group since those work pools were not cloned). 3) If an entry was previously marked as the default work pool, the newly cloned workpool will be marked as the default workpool. (This is a change in logic, as previously the access group always used the default work pool [Org-App-Work]). 4) If the previous default work pool was not cloned to the new application (based on case type selection), then the access group's default workpool will be the first in the list. 5) If there were no work pools present in the list, then the standard Org-App-Work classgroup that is always created as part of the new application process will be added and made the default.
SR-B36689 · Issue 292880
Performance improvements for CaptureInteraction without VBD
Resolved in Pega Version 7.3
CaptureInteraction was slow when not using a VBD server due to a call to the function GetVBDService that fired even when there was no intention to use VBD and it was not configured. This was an unintentional side effect of code that listens for VBD being ready in case it is in mid-initialization or cluster nodes are restarting. To improve performance, GetVBDService has been modified to avoid waiting for the VBD cluster if VBD is disabled using DSS prconfig/initservices/vbd/default = false.
SR-B36743 · Issue 293937
BLOB data save performance improved
Resolved in Pega Version 7.3
In an installation with many large work objects, performance issues were seen when saving a BLOB. This was due to the BLOB still saving data for properties that had been removed, and has been resolved through better tracking of orphan entries of list type properties like string list and java object list.
SR-B36880 · Issue 294350
Fixed SmartInfo positioning
Resolved in Pega Version 7.3
When any action like Refresh section, Post Value, or set value was configured before the Show SmartInfo, the smartinfo was suppressed. This was because the TargetElement was coming as stale and positioning the smartinfo based on a stale element resulted in it not getting displayed. This has been fixed by adding code to update the latest target element before positioning the smartinfo.
SR-B36881 · Issue 296927
ADM service performance improvements
Resolved in Pega Version 7.3
The memory usage and allocation logic for ADM has been tuned for performance improvements.
SR-B36881 · Issue 301076
ADM service performance improvements
Resolved in Pega Version 7.3
The memory usage and allocation logic for ADM has been tuned for performance improvements.
SR-B36909 · Issue 301697
JSON API added
Resolved in Pega Version 7.3
An enhancement has been added via an API to create JSON from clipboard page. This API will get the JSON of the clipboard page, expand Autopopulate Properties, and return Dates and Times in ISO format.
SR-B36948 · Issue 296463
Dynamic UI supports repeating grid sourced from Page List
Resolved in Pega Version 7.3
After creating and running a case, clicking on the "Edit Details" option from Other Actions Menu did not display the section containing a repeating grid sourced from a Page List. This was due to the final rule activity "pzdynamicuisetupfields" not considering the case of Page List while configuring the dynamic UI, and it has been updated to properly propagate the values.
SR-B36948 · Issue 298094
Dynamic UI supports repeating grid sourced from Page List
Resolved in Pega Version 7.3
After creating and running a case, clicking on the "Edit Details" option from Other Actions Menu did not display the section containing a repeating grid sourced from a Page List. This was due to the final rule activity "pzdynamicuisetupfields" not considering the case of Page List while configuring the dynamic UI, and it has been updated to properly propagate the values.
SR-B36974 · Issue 293707
VBD improvements and fixes
Resolved in Pega Version 7.3
The following issues have been addressed: 1) VBD was generating the message "Error converting DB record to VBDRecord" when IH contained 'timestamp with time zone' fields and the backend was Oracle. This has been resolved by modifying the Partition loading logic to support Oracle Timestamp type fields. 2) VBD was loading data from all columns in the IH tables, including those not needed in VBD. To improve performance, the Partition loading logic will filter out IH DB fields that are not set up in IHConfig. 3) The logic for unpacking Cassandra (for access to diagnostic tools) into the app server directory would fail without proper permissions. To avoid this Cassandra will be unpacked into the Pega Platform work directory instead of the app server directory. Work will not need to be cleared as this Cassandra directory is for access to diagnostic tools only. 4) Eager loading will indicate in the logs when the load is finished with the message "Eager mode loading of partitions completed. X partitions(Y rows) loaded." 5) DB check using min/max query could be slow when given a large fact table, leading to data flow timeouts. To fix this, DB check will no longer occur on the same thread that processes Data Flow inserts. Additionally, the DB check will only scan for min instead of min/max. Operations that do require min and max have been split into two queries (one for min, one for max). 6) Previously, if a VBD Client node was disconnected from the cluster due to Pulse not finding any VBD server nodes, only a VBD server node restart or VBD Client node restart would get them reconnected. To make this process easier, VBD Pulse will reconnect the client if it is found to be disconnected while the VBD cluster is up.