The availability of the following new features, enhancements and corrected issues is dependent on the installed edition of the product and licensed features.
ERP nodes to handle the max row count
The Extract ERP Table node will allow you to set the MaxRowCount that limits the number of rows retrieved when no RowCount is specified.
The RowCount property is used to limit the rows, however, if the SAP system does not allow for the default (2147483647) number of rows, the property can be "unhidden" then set on the node or set server-wide via adding a line in the cust.prop to: ls.brain.node.erp.sapconnector.extractTable.maxRowCount=<maxRowCountValue>
Preserve Tomcat settings across upgrades
The upgrade installation process has been improved by ensuring that server.xml, web.xml & cacerts keystore are maintained between upgrades. Examples of when you might change these files are while Configuring HTTPS or while Importing SSL certificates for use with certain nodes.
Note: For 3.10.1 the changes will still need to be made to the files, they will be preserved after future upgrades. Also the instructions for making such changes have changed compared to previous versions.
Other changes
- Fixed an issue where the Extract ERP Table node sometimes fails with an error "Option not valid" as the OPTIONS clause generated by the node was not always inserting the correct spaces surrounding parentheses.
- The Database Metadata node has been changed to make trigger queries against Oracle using DBA_TRIGGERS instead of ALL_TRIGGERS. This should remove the need for any trigger permissions via "grant create any trigger to <user>" and only select permissions should be required via "grant SELECT_CATALOG_ROLE to <user>".
- Introduced an ability to disable the schedule for a data flow while it is running.
- It is now possible to download the node logs directly from the canvas using the download option available when you right-click on the respective node.
- Introduced fistExec, lastExec, and execCount functions to the advanced scripting capability of the Lookup node.
- Introduced a new property in the Execution Sandbox. ls.brain.server.docker.mounts allows the user to configure access to file locations other than the locations restricted to the Execution Sandbox.
- Fixed an issue whereby if multiple executions started simultaneously as the first runs after a server restart, the executions could result in error stating "Multiple elements are in the collection" due to a race condition. This error would then reoccur on all executions until a subsequent restart.
- Added a new property "
CreateFileUsingAbsolutePath" to the "Archive Compress" node, where the backwards compatibility is set to true. When set to false the path of the files in the created zip file will be relative to the source folder, as opposed to using the full file path of the source system which previously caused issues with the default Windows zip tool when trying to open the zip files produced by the node.
Corrected issues
|
Issue summary |
Issue number |
|---|---|
| Fixed an issue in the Database Metadata node for DB2, the Procedure Parameters output had a field called TABLE_NAME which should have been called PROCEDURE_NAME. On the Function Parameters output, had a field called TABLE_NAME which should have been called FUNCTION_NAME. | LAE-28804 |
| Fixed an issue whereby the d3sa.api.rest.host or ls.brain.webapp.contextUrl were not used when creating the URL for the URL_To_Linked_Run property in the failed schedule e-mail notification. | LAE-28819 |
| Fixed breadcrumbs on upload file to server dialog. | LAE-28927 |
| Fixed an issue where toggle temporary transparency was not working for textbox. | LAE-28784 |
| Fixed an issue in URL Query Builder Node whereby fields specified Excluded Fields property were not excluded. | LAE-28948 |
| Fixed an issue where in the Database Metadata node, the table columns, and procedure and function parameters outputs previously had a null catalog field when run against postgres. While postgres does not return this information, the node knows which catalog it is working against to include the catalog information. | LAE-28966 |
| Fixed an issue where in 3.10.0 while running the Database Metadata node against an Azure Synapse database, if there were more than 32 thousand columns across all tables in any single schema, the node would log a warning and not retrieve any of the columns for that schema. | LAE-28941 |
| Fixed an issue whereby services does not initiate when installing with restore from backup option. | LAE-28774 |
| Fixed an issue where system backups could occasionally result in the application winding up in a deadlock requiring an application restart. | LAE-27566 |
| Fixed an issue when an Execute Data flow node is running while a system backup is ongoing, it will not be able to finish until the backup is complete. | LAE-26357 |
| Fixed an issue where system backups could occasionally result in the application winding up in a deadlock requiring an application restart. | LAE-27690 |
| Fixed an issue where logged errors are not displayed if a node is completed successfully. | LAE-28619 |
| Fixed an issue on the HTTP node where the DataOutputFieldType property was not always being used if one of the other output field properties were set. | LAE-27559 |
| Fixed an issue where multiple 'do output' statements in a single brainscript script pointing to the same output were not always writing the records correctly. | LAE-28613 |
| Fixed an issue where the file picker does not show the files in the directory when the file picker is re-opened following the selection of a file in a sub-directory. |
LAE-28852 |
| Fixes an issue whereby the Database Metadata node was reporting warnings about "USE statement is not supported" and causing the node to not retrieve correct information when running against an Azure Synapse database. | LAE-28939 |
| Fixed an issue where the Database Profiler node will report the internal errors that occur during the column analysis phase of the profiling to the node's error output but this will not cause the node to fail unless the FailureBehavior property is set to Error. | LAE-28875 |
| Fixed an issue in the Excel File node if a WorkbookSpec property was specified which did not have the start and end column defined, and the node was configured to DetectFieldTypes, and the header row in the Excel spreadsheet contained empty cells, the cells were not getting auto-named correctly and instead were winding up with field names from the first row of the data. | LAE-28351 |
| Fixed an issue whereby it was not possible to rename the output pin on the Modify Field Prefix node. | LAE-28615 |
| Fixed an issue where the Query for Salesforce node would show sensitive information when logging errors. | LAE-28991 |
| Fixes an issue whereby users with long user names could not execute data flows in the Execution Sandbox configuration of . | LAE-29196 |
| Docker network was removed after restart of Windows VM. This issue has been changed such that by default, on windows, when the server starts up and is configured to use execution containers, the named network will be recreated if it does not exist. | LAE-28936 |
| Fixed Salesforce node where Id mapping was incorrect for insert operation. | LAE-28808 |
| Fixed an issue whereby the Modify Field Prefix node was not adding the prefix in a case-sensitive manner. | LAE-28782 |
| Fixed an issue in the Convert Tabular XML node, where the node was incorrectly dealing with the 'From Field' choice, instead of referencing the specified input field. Hence, the node was processing the fieldname as literal xml and throwing a parsing error. | LAE-24871 |
| Fixed an issue where the Query for Salesforce node would fail on some unicode characters. | LAE-27674 |