INC-166498 · Issue 637054
Performance improvements for sites with a large number of tables
Resolved in Pega Version 8.3.6
Performance impacts were seen when there were a very large number of tables in the database and the cluster/node persistence jobs were running. This was traced to the volume of metadata / schema introspection queries fired from the pyPersistNodeState and pyPersistClusterState jobs along with the Classmap cache invalidation / lookup queries, and has been resolved by improving the code to avoid database introspection when the current number of tables is greater than a maximum value.
INC-166631 · Issue 637021
Optimizing helper class enhanced to handle external databases
Resolved in Pega Version 8.3.6
Running a BIX extract that included a manifest for a target database was resulting in a null pointer exception for the manifest extraction. Attempting to generate the DDL for the manifest table also failed. This was traced to an issue with the helper class using a hardcoded default database for forming the queries, causing it to ignore the DB config/DADN/prconfig for the Oracle database and form a query using the PegaRules' database credentials. This only occurred when trying to do external database operations on a different DB platform; Oracle PegaRules worked as expected with an Oracle external DB and Postgres Pegarules worked with a Postgres external DB, but mixing Postgres PegaRules and an Oracle external DB would result in the null pointer error. To resolve this, the helper class has been enhanced to work with external databases by passing the database name as a parameter so it will properly calculate the query based on the type of target. The default for 'database/oracle/allowOptimization' is set to false: additional database privileges must be added or some configurations must be updated to provide the information needed in lieu of the privileges before making use of these optimizations. In addition, an error in the name of the class has also been corrected and is now available as PerformanceHelper rather than the previous "PerformaneHelper".
INC-168334 · Issue 648372
Locking added to avoid null pointer error for auto-populate property
Resolved in Pega Version 8.3.6
After configuring the auto populate property "OrgProduct" which referred to a data page, the system experiencing heavy load led to the property not getting properly initialized. This resulted in a WrongModeException and NullPointerException. To resolve this, the system has been updated to lock the requestor when Queue Processors execute their activity. This will prevent race conditions and concurrent modifications if other threads are accessing the same requestor.
INC-169764 · Issue 645803
Decision table query performance improvements
Resolved in Pega Version 8.3.6
Severe latency was seen after upgrade, and it was not possible to save business rule changes. This was traced to an issue with the decision rule being treated as a custom rule, and has been resolved by inserting a check. In addition, optimization has been added for queries on the pr_sys_decchg table.
INC-170841 · Issue 646186
Resolved PRQueueLockException
Resolved in Pega Version 8.3.6
A Class Queue not found exception resulted when an SLA was deleted on the same transaction where it was created when deferred save was used. This was an inadvertent side effect of work done on the QueueManager implementation to invalidate previously deferred operations for item and schedule removal, and has been resolved by clearing the processing state for PRQueueLockException when an item reservation fails.
INC-143121 · Issue 610732
Timeout for loading predictors made configurable
Resolved in Pega Version 8.3.6
When using an extremely large number of predictors, the Report definition pzADMPredictorsFilter was suffering timeouts due to the time for loading predictors from the database exceeding the time threshold allowed. This has been resolved by marking the rule as editable to allow custom setting of the threshold according to need.
INC-148899 · Issue 615702
Adaptive models update correctly
Resolved in Pega Version 8.3.6
Some models had the recorded responses column updated, but the models (number of Positive, Negative and Processed Responses) were not updated. Investigation showed that deleting the modelRuleConfiguration through the stateManager/client did not delete modelFactories related to the configuration. If a new configuration came in with a different algorithm, the update issue occurred. This has been resolved by reseting the configuration according to its factory in that specific case.
INC-155822 · Issue 618267
Locking added to avoid null pointer error for auto-populate property
Resolved in Pega Version 8.3.6
After configuring the auto populate property "OrgProduct" which referred to a data page, the system experiencing heavy load led to the property not getting properly initialized. This resulted in a WrongModeException and NullPointerException. To resolve this, the system has been updated to lock the requestor when Queue Processors execute their activity. This will prevent race conditions and concurrent modifications if other threads are accessing the same requestor.
INC-157629 · Issue 626633
Duplicate key exception resolved for adaptive model
Resolved in Pega Version 8.3.6
During the model snapshot update, a DuplicateKeyException was generated while trying to insert a record in to the predictor table. This did not affect the model's learning, but did appear ion the model monitoring report. This was traced to a local scenario of having the same outcome values defined on the model with different cases (Accept and accept). All predictors used in an Adaptive model are inserted into the model monitoring tables as a part of the monitoring job: because the monitoring tables are not case sensitive, this lead to a unique constraint exception since there were multiple IH predictors with the same name. To resolve this, validation has been added which will skip adding duplicates from new responses.
INC-160331 · Issue 628710
ML model continue tag error changed to debug logging
Resolved in Pega Version 8.3.6
Log entries were seen indicating "The continue tag is found without a start tag for window Line1 1234567890 in text". This appeared to be coming from the MLCommand java class. Investigation showed this occurred when the ML model believed a portion of the text (token) was a continuation of the entity because it has not found the starting part of the entity. This situation is not currently treated as a valid output, so an update has been made to change the logging level from error to debug for this case.