SR-D90687 · Issue 560431
IOException handling improved to resolve broken pipe errors
Resolved in Pega Version 8.4.2
Frequent "connection reset by peers" exceptions were being generated and broken-pipe exceptions were seen in the logs. Investigation traced the issue to unhanded IOExceptions on the server side that were a result of the client application not always closing the TCP connection gracefully. To resolve this, error handling for IOExceptions has been improved.
SR-D67408 · Issue 554900
Directory traversal blocked in zip import
Resolved in Pega Version 8.4.2
One of the files contained in a zip archive was not deleted from the system after zip import. This was due to the file being created by a third-party archive that included of a directory traversal character that caused it to be inflated outside of the temp directory. To resolve this, a check has been added to that a file with directory traversal characters in its name will not be inflated.
SR-D81572 · Issue 551028
JDBC URL handling added for Oracle over TCPS
Resolved in Pega Version 8.4.2
While attempting to upgrade an environment over TCPS, the generateDDL.sh script was failing. The same environment ran without issue on Tomcat with the same URL. Investigation showed the JDBC url was not correctly generated while running the upgrade: in a standard scenario, there will be no spaces in the JDBC URL specified. However, because Oracle can send spaces as part of JDBC URL and cause this issue, an update has been made which will quote the JDBC URL argument for the ant target in setupDDL.xml.
SR-D84364 · Issue 551403
Check for circular references added to SearchInventoryImpl to prevent recursive call
Resolved in Pega Version 8.4.2
An out of memory error was traced to SearchInventoryImpl infinitely recursing over a clipboard property, where the child property referenced a parent property and resulted in an endless loop. This has been resolved with the addition of a depth check to ensure that the search does not recurse infinitely.
SR-D85100 · Issue 556262
ProductInfoReader updated to fetch only most recent version information
Resolved in Pega Version 8.4.2
After upgrade, running Hfix scanner on Pega Marketing 8.2 displayed missed critical Hfixes for Pega Marketing 8.1. This has been resolved by modifying ProductInfoReader.runQuery to fetch only latest version of DAPF instances during a scan.
SR-D98404 · Issue 558207
Handling added for hotfix Rule-Application instances
Resolved in Pega Version 8.4.2
A null pointer error during DL file expansion performed as part of the second phase of a hotfix installation caused the hotfix install to fail. The null-pointer exception was thrown because the code, primarily used for export, performed a database lookup of a Rule-Application and assumed the response would be non-null without checking the result. During export the Rule-Application would normally exist because the system would have interacted with it already during the export by identifying it and writing it to the archive. During phased hotfix installation, rules are staged to the database in a different table during the first phase and reconstituted during the second phase. The scenario for this error was a missed corner case specific to the unusual combination of including Rule-Application instances in a platform hotfix. To resolve this and prevent further issues, handling has been added for this use case.
SR-D20439 · Issue 496405
Hotfix hashmap cleanup improved
Resolved in Pega Version 8.1.7
A DL installation failed with an out-of-memory error. It was observed that many CacheEntry(VersionedJdbcJarRwader.java) hashmaps were being created to store the details for each hotfix but that were then never cleared. This has been resolved by adding the finally block in CodeImportProcessImpl.java class to call the shutdown method for proper cleanup. In addition, the default command line utility (prpcUtils) has been updated to use JVM settings of Xmx 4GB and capture a heap dump if an out-of-memory error is encountered. The auto-generated prconfig.xml has also been updated to leverage its minimal startup setting which will avoid loading the conclusion cache into memory.
SR-D62754 · Issue 559848
PrepareResponse updated to explictly close Input Handler
Resolved in Pega Version 8.2.7
When using prpcServiceUtils to export a product in a Windows+Weblogic environment, attempting to export repeatedly using the same archiveName with the intention of overwriting the older product with the newer one in the ServiceExport directory failed with a FileNotFoundException. Investigation showed that the product file that was created by the pzExport REST call was not being released by the Weblogic File Handler process. Due to this, the next time the call was invoked the system tried to create the same file in the directory but failed due to the earlier File handle lock. To resolve this, the system has been updated to explicitly close the InputStream using try-with-resources.
SR-D78274 · Issue 544092
Handling added for dual privileges with MSSQL
Resolved in Pega Version 8.2.7
After setting up dual privileges, the Admin user was able to create a table but the base user received an "insufficient privileges" error. Investigation showed this was an issue when using MSSQL: the generated grant statements used the server login name as the user in the grant statement, instead of the database user. For all other databases, the username passed into the connection is the correct user to use for grants. Only MSSQL has a distinction between this connection user name (the login) and the database user, and since the login did not exist in the user table, the grant failed. To resolve this, when MSSQL is used, the system will fetch the underlying database user when determining the user for grant statement generation.
SR-D84190 · Issue 547173
Post-Import Migration Agent query optimized
Resolved in Pega Version 8.2.7
A Post-Import Migration agent belonging to the Pega-ImportExport Ruleset and set to run every 60 seconds by default triggered the SQL query "select ASTERISK from pegadata.pca_CWT_CXP_Work_Interaction" which ran for an excessive amount of time, caused a utilization spike, and then crashed the utility nodes. Investigation showed the excessive run time and load was caused by the query fetching a very large number of results. To better handle this scenario, the query usage has been updated.