INC-192025 · Issue 680707
Scenario Planner performance improvements
Resolved in Pega Version 8.6.3
Scenario planner runs were causing out of memory errors on the dataflow nodes. To resolve this, updates have been made to improve performance. Only relevant information will be kept for decision funnel simulations out of SR pages, the memory footprint of data structures used to aggregate and rollup counts has been reduced, and temporary objects will be released as soon as they are not needed anymore.
INC-192102 · Issue 676361
Performance improvements for stale thread warnings
Resolved in Pega Version 8.6.3
Stale thread warnings were causing performance issues during dataflow run execution. Stale thread/slow component warnings are added as part of the dataflow execution when a processing thread takes more than 5 minutes to process a single dataflow record. The stacktrace of the dataflow thread is added as part of the warning for debugging purposes, but in some scenarios the stacktrace can become very large. This has been resolved by removing the stacktraces from the warning, improving the query logic, and adding the run ID to the exception method to assist if there is an error.
INC-193399 · Issue 688112
DSS added to handle merges with lower versions of Postgres
Resolved in Pega Version 8.6.3
After update, executing the batch campaign with volume constraint resulted in the second data flow DF_Wait failing with error message "ERROR: number of columns (1844) exceeds limit (1664)". This was due to the database set’s change (in 8.5) to use the database layer’s merge statement. Prior to that, the logic used "deletes and inserts". Depending on the version of Postsgres, the generated SQL statement for a merge statement is different. The “INSERT … ON CONFLICT … UPDATE” syntax is generated for Postgres 9.5+ AND when there is a PK constraint defined for the DB table. Otherwise, the complex UPSERT statement (old syntax) is generated, as was the case in this issue. This is a known issue in the Postgres server software where it mis-interprets the number of columns involved. i.e., it mistakenly counts the number of columns twice. As a result, the actual maximum columns allowed is only half of the official limit (1664). The same UPSERT statement does not cause the “exceeds limit” exception if there are 832 or fewer columns in the statement. To resolve this, an option has been provided to select between the “original logic” (deletes and inserts) and the “merge statements” logic by way of the DSS “decision/datasets/db/useMergeStatementForUpdates”. Setting “true” will use the merge statement logic, and setting “false” will use deletes and inserts. When the DSS is not defined, the default is "true" and the system will use merge statements in the form preferred by Postgres 9.5+.
INC-193632 · Issue 679173
Cassandra driver metrics exposed for performance troubleshooting
Resolved in Pega Version 8.6.3
By default Cassandra driver metrics are now enabled. Metrics can be disabled by setting the dnode/disable_driver_metrics prconfig parameter.
INC-193847 · Issue 695975
DSS added to allow masking of subjectID in alerts
Resolved in Pega Version 8.6.3
In order to allow customizing whether or not a subjectID is included in alerts, a DSS has been added to conditionally mask the subjectID from being logged. To use this, set the "alerts/maskIHsubjectID" DSS in the Pega-DecisionEngine ruleset to true to hide the pySubjectID.
INC-193986 · Issue 680033
Parameter logic updated for metrics activity counter
Resolved in Pega Version 8.6.3
An error was causing the PushCDHMetrics agent to fail. This was traced to an undefined parameter in the activity which was used as a counter, and has been resolved by replacing it with a local variable of type integer.
INC-194305 · Issue 681682
File Dataset wildcards updated for ADM storage
Resolved in Pega Version 8.6.3
Attempting to access the filepath was failing and an error was seen indicating "Could not obtain lock 'Create repository file pegacloudfilestorage:ADM/Rule-Decision-AdaptiveModel/Data-Decision-Request-Customer". While creating a file name, a lock is generated while checking the file names of existing files. If the file data set has a large number of files, a condition may occur where some threads are unable to save data and an exception is thrown. To resolve this, the wildcard previously used in the file name has been replaced with the Pega node id + thread id + current timestamp. This will ensure all wildcards are unique and there is no need to lock and list existing files.
INC-194405 · Issue 680278
Performance improvements for globally optimized strategies
Resolved in Pega Version 8.6.3
After update, the GOS_NBA_TopLevel strategy was consuming a large amount of processing time related to the hashmap. This was traced to the lookup of a value in a set of Class<?> that used Object.HashCode. To improve performance, the set lookup has been removed and replaced with checking validity by the type of an object.
INC-194433 · Issue 679631
Key Value Store handling revised for greater backwards compatibility
Resolved in Pega Version 8.6.3
Previous optimization work done to reduce the size of the "nm" field when the hierarchy string is empty resulted in some Key Value Store records not being read from older systems during update. This has been corrected.
INC-194657 · Issue 682006
ADMinputsource field population updated to handle transactional decisions
Resolved in Pega Version 8.6.3
The CaptureResponse flow was failing while writing to AdaptiveAnalytics with the error "IllegalArgumentException - argument does not represent JSON object". This occurred while running an outbound campaign where a few decisions were treated as transactional and did not have Model executions. Attempting to set a response for these decisions resulted in a JSON parse exception being thrown due to pxADMInputs not being populated (no models executed). This occurred only when using transactional actions in CDH, and was caused by the system only including predictive models during the make decision flow due to common inputs not being stored. While there was a workaround of overriding the property pzADMInputSource to use modelReferences instead of admInputContainer, this has been resolved by correcting how the pzADMinputsource field is populated. When there is no modelInput, the system will not populate that field so the page is ignored.