New Features
Glue Queue Overview
New screen is available that provides you an overview of existing Glue Queues.
The screen shows various queue parameters, including the number of subscribers, number of packages, and size of data stored in a queue. This provides you with detailed information on the usage and the current situation of Glue queues in the SAP system.
Improvements
SLT Integration
The implementation of SLT integration has been changed so that the table structure is read directly from SLT metadata. In the previous version, this was obtained from the source SAP system using an RFC call. The drawback with the original approach was that additional authorizations to a different system were required and also that SLT does not always replicate the original structure of the source table.
Glue provides now a safety mechanism that prevents the accumulation of large amounts of data in Glue queues if e.g. target storage is not accessible or if the extraction watchdog is paused. With this enhancement, the SLT replication is paused after data volume in Glue queues reaches the critical limit.
Glue metadata now provides information about SLT load type. This gives you the possibility to distinguish full and delta load when further processing the replicated data on the target storage.
Extractors
BEx Query (BICS) fetcher filter now supports selections on Hierarchy Node variables and Hierarchy variables. Also, the screen of the BEx Query (BICS) fetcher has been adjusted so that it is more usable on smaller screen resolutions.
Glue request monitor now provides additional information about the date and time of replication.
It is no longer possible to execute the extraction with load type Delta Load and Delta Init Without Data using Mass Execution.
Glue extraction scheduler provides an option to save variants of the created configuration.
When executing an extraction from the SAP table, you can specify Database Hints. This enables you to set DB-specific instructions to e.g. optimize performance.
HANA Native extraction
In HANA Glue Fetcher, you can use HANA Input Parameters for filtering the data that will be replicated.
TMS
In Glue TMS you can now filter Glue objects by name.
Glue Profiles
In Glue table File Options, if you are using File Storage profiles, you can specify a custom container name and hence overwrite the profile settings.
New placeholders were added to Glue profiles that enable you to add change time, change date, and system ID tags to generated file names.
Additional features
Glue screens have been optimized for SAP Belize and Quartz themes. Also, screens of Extractor 2.0 objects have been reorganized to display more information.
Storage Management - Snowflake
For the connection to Snowflake, you can now use Key pair authentication. Both encrypted and unencrypted versions are supported.
In the previous version, you had to specify authentication parameters for both Snowflake Internal Stage and Snowflake storages. The configuration has been simplified and now you define these parameters only in Snowflake internal stage.
Repeat parameters for the Snowflake connector have been added to deal with unstable network connection situations.
You can transfer data to Snowflake using parquet files.
You can use Snowflake DB hints.
Parquet file support
When replicating data to parquet files, you can specify how to handle inconsistent date values. In the SAP system, date values are stored in various formats which may not always be compatible with the target technology. A new setting has been added that gives you the possibility to choose if, in case of incorrect value, the extraction fails, or empty values are inserted and the procedure continues.
Fixes
Data Dictionary
It was not possible to create a Glue table from the external table with columns containing ‘-' character or empty space ' '. New logic has been adjusted to replace such characters with '_’ so that SNP Glue can handle also cases like this.
Extractors
Extraction Process failed to generate correct mapping if there were multiple fields whose names differed after the 23rd character.
Extraction Process filter selection was not supporting String data elements. Also, it wasn’t possible to use fields containing the character ' in their name for filtering.
It was not possible to activate the Extraction Process when the field name had the same name as the table.
Editing of Glue Fetcher's description was prompting for the deactivation of an object. This has been corrected and you can change the description of a fetcher without deactivation.
The extraction from CSV file to target storage has been improved. The performance and stability of the solution have been enhanced.
When extracting the data to files, the generated JSON metadata contained incorrect information about request status and finish time. This has been fixed and correct data is now filled.
In the Extractor manage screen, extraction jobs were not displayed when the SAP system and user computer were in different time zones. The logic has been adjusted to display jobs correctly.
Glue Queue
After the deletion of a Glue Queue, metadata was not cleared properly.
Queue watchdog was in special cases running the extraction in parallel even if the parallelization was disabled.
SLT
If the Glue table was created automatically using SLT, the column description was missing in metadata. This has been corrected and the Glue table is generated properly now.
When an object is added to the SLT replication scenario, all necessary Glue objects are generated and the newly created Extraction Process is added to the Queue watchdog. The problem was when this process was later removed from the watchdog for manual execution, with the next SLT execution it was again assigned back to the watchdog. This has been fixed and the process is assigned to Queue watchdog only once, right after it is created.
CDC
Trigger delta was not working properly when used on a cluster table with multiple cluster pages. The logic has been enhanced to cover also such cases.
Deletion flag was not filled correctly in case the trigger delta on the cluster table captured more records than could fit into a single package.
Extraction Scheduler
Job name and Extraction Process variant were not saved in the variant of an Extraction Scheduler. The logic has been corrected and now all the settings are properly stored.
Request Monitor
In the Glue request monitor, it was not possible to specify a selection using String fields.
The performance of the monitor has been improved to handle large volumes of entries.
ODP
ODP fetcher was not able to reset delta if no ODP queue subscriber was found. This resulted in a failing extraction job. The logic has been adjusted to skip delta reset in such a case.
TMS
Descriptions of Glue Fetcher and Consumer were not transported using Glue TMS.
Logical folders were not transported. This has been fixed and you can transport logical folders as well.
BEx Query (BICS) Fetcher variants were not transported correctly.
Additional features
Modifiable system check was applied to the activation/deactivation of Glue objects. With the latest version, the activation and deactivation are allowed and a modifiable check applies only to edit action.
In case Mass Object Generation was executed with an empty Target object ID, only Glue tables were created without additional Glue objects.
Unused parameters have been removed from the Mass Process Regeneration screen.
Glue cockpit dashboard has been reworked to be able to display information on huge numbers of requests.