(Glue-2305) Release Notes 2305

New Features

Storage Management

You can now connect to Azure Blob Immutable storage as a target in Storage management and thus store business-critical data in a WORM (Write Once, Read Many) state.

Improvements

Glue Scheduler

In the current release, Trigger and Queue watchdogs, and Extraction Scheduler have been removed from the SNP Glue cockpit. The individual tools are still supported and maintained and they can be accessed from the menu Glue Cockpit > Tools > Glue utilities.

A new tool called Watchdog Migration has been implemented to support a smooth transition from the old watchdogs and extraction scheduler to the new Glue Scheduler. Please, use the following document for more information.

SLT Integration

• You can now use SLT also for replication of non-ABAP sources, e.g. Oracle tables. Previously, only standard ABAP source objects were supported.

Extractors

• Listcube extraction now supports mass execution. A newly implemented partitioning logic splits the source data automatically based on selected InfoProvider characteristics. This data split allows extracting the whole InfoProvider data in smaller portions thus avoiding memory overflow issues.

• Mass execution for BEx Query (BICS) has been improved as well by adding the new parameter Number of portions.

• Several usability improvements have been implemented for Glue Scheduler. When selecting extraction processes and their variants for the scheduling, you can now use F4 help to select a variant.
  Also, parallelization is now supported for the batch execution of the scheduler of the type queue.

• The (Glue-2305) Glue Queue screen has been enhanced by providing additional information about the age of cluster IDs stored in the selected queues.

• GLUE-4662 Read all clients in the fetcher causes DBSQL\_ILLEGAL\_CLIENT\_SPECIFIED exception

CDC

• The performance of trigger delta buffering has been improved significantly.

Storage Management

• You can now configure a connection to Azure Data Lake Gen 2 storage without using the SAP ICM component. Please, use the following documentation for more information.

• An optimization has been implemented for BLOB DB connections, which improves the performance of the read operation. The improvement is available for the following storages: MaxDB BLOB MS SQL BLOB Oracle BLOB SAP ASE BLOB, and BLOB Storage (Primary Database).

Additional Information

• Glue can now be installed also on SAP GRC systems.

Fixes

Data Browser

• Glue Data Browser didn’t display any data if the filter selection contained a space. This has been corrected and now you can filter based on values containing space characters.

Extractors

• SNP Glue was able to read data from multiple clients only in case the table contained the MANDT field on the first position of a table. More complex logic has been implemented to enable this scenario for all tables that have on the first position of the table a datatype CLNT with length 3 which is part of the primary key.

• After the last release, filters were not visible in Fetcher objects anymore for ADSOs. This has been corrected and now you can use ADSO fetcher without any restrictions.

• When running mass execution, the delta init phase wasn’t in some cases executed properly. The logic has been corrected and now the whole execution runs properly.

• The extraction using BEx Query BICS was not working when a field was imported from a BEx query and a characteristic had a Display setting set to “No Display“.

SLT

• When adding a large number of tables into SLT, some weren’t saved into Scheduler. The logic has been corrected and all tables are now added to the scheduler correctly.

Glue Scheduler

• In the Glue Scheduler, you could add an extraction process and specify a non-existing variant. With the latest release, you can specify only existing variants.

• The scheduler allowed you to add an inactive extraction process to your configuration and schedule it. In the new version, this is still enabled, but a warning message will appear to make you aware of inactive processes.

• If an extraction process variant was executed using the Glue scheduler, the variant name was not displayed in the extraction monitor.

• It was not possible to specify a package size in the Trigger type of the Glue Scheduler.

• The status of an extraction process was not updated in the Glue Scheduler after the extraction job ended with an error.

• The Glue Scheduler started extraction from ODP with a full load even though a delta load was selected.

CDC

• It was not possible to create a trigger in the MSSQL database if the database table name contained a “/“ character.

• A Glue trigger was not updated after the source table structure was altered. This resulted in replicating an incomplete data set. The logic has been adjusted and now the trigger structure is updated and the correct data transferred.

Storage Management

• Azure Delta Lake metadata was not consistent after a checkpoint on the delta table was executed. Also, the parallel write into the delta table was not working correctly. Both of the issues have been resolved and now the data is correctly inserted into the delta lake table.

• It was not possible to change the RFC connection in the Snowflake storage.

• The communication between Azure Blob and Azure SQL was not working if the SAS token of the Azure Blob storage was hashed.

Known issues

• When replicating data from Glue queues from another SAP system using an RFC connection, a streaming-optimized Queue Fetcher is not working properly. It is recommended to use a standard Queue fetcher instead. Please contact us in case you need more detailed information on how to configure such a scenario.

• SLT integration in the Separate box deployment (SLT and SNP Glue run on separate SAP systems) is under certain conditions malfunctioning. Please, do not upgrade to this version in case you use this scenario.

• In case you use Trigger delta on a View that has different key field names than the source table, and these fields are defined by data types, Trigger activation will fail.