(Glue-2211) Release Notes 2211

Owned by Igor Strba
Last updated: Mar 27, 2023 by Robert Adamec
4 min read

New Features

Glue Central Instance

Glue Central Instance is used for the remote administration of Glue processes. It displays an overview of Extraction processes, Glue queues, Glue watchdogs, and requests.
Based on advanced filtering, it is possible to view created Extraction processes on the remote system and then run the extraction with the selected variant.

Glue Scheduler

A Glue Scheduler is a new tool for scheduling Glue extractions and is meant to be a central place for scheduling and management of all the replication jobs.
In the future, it will replace existing Trigger and Queue watchdogs because it overcomes several limitations of the existing functionality.

In the current version, Glue Scheduler supports the scheduling of extractions from Glue Queues. You can create your own configurations, specify parameters like the number of jobs, parallelization, and type of extraction (whether it is executed as a batch job or as a streaming replication), and choose extraction processes that will be executed as part of the configuration.

Storage Management

You can now connect to SAP ASE BLOB as a target in Storage management 

Improvements

SLT Integration

  • SLT implementation now supports delta replication on cluster tables.

  • The SNP Glue SLT integration module now enables you to create a table (in case of extracting data to a table) or container (extraction to files) with the same name as the SAP source table. Even lowercase names are supported.
    This can be accomplished by using Glue Profiles. Please note that these parameters are available also in SLT settings.

  • You can schedule the replication from SLT using the new Glue Scheduler.

  • Glue Scheduler gives you more control over the replication when scheduling a combination of Full and Streaming replication modes. This feature can be used also in combination with SLT.

  • When replicating the data using SLT, all Glue technical fields are generated at the end of a table structure, unless these are set as key fields.

Extractors

  • BICS Fetcher now supports extraction from BEx Queries with a dynamic selection of a hierarchy.

Storage Management

  • Storage Management now stores only hashed passwords. The specified password is automatically decoded using a Master key of your choice.

  • Default mapping for Date and Time values has been implemented so that such values are automatically mapped to corresponding Google BigQuery data types.

Fixes

Data Dictionary

  • When importing fields from a table that contains a RAW key field into the Glue table, the field was changed to String and the key field flag was removed. The behavior has been changed so even though the field is changed to String it stays as the key field.
    For storages that don't support strings in key fields, an error message is displayed when activating the table.

Extractors

  • The extraction process Manage screen didn’t display selection parameters when an ABAP fetcher was used.

  • Mass execution logic that was splitting data into ranges was not respecting lowercase values.

  • The generation of extraction metadata files has been changed when replicating the data using mass execution with the truncate option enabled. Right after the mass execution is scheduled, a new JSON file is created which contains a truncate flag set to "X" (enabled). The JSON files for individual extraction jobs contain an empty truncate flag.

  • Queue fetcher optimized for streaming data replication in rare cases didn’t transfer all the data to the target storage. This has been corrected and the fetcher works properly, now.

  • If there were too many small packages stored in a Glue queue, the replication was failing because the generated SQL statement was too big.

CDC

  • In the previous release, it was possible to run a delta watchdog in streaming mode with full loads and with mass execution. This could potentially lead to millions of generated jobs. The logic has been adjusted to prevent such situations.

  • Trigger watchdog sometimes scheduled 2 extraction jobs. The logic has been adjusted and works correctly now.

  • The deduplication of the DSO delta was not working correctly. In the new release, the logic has been corrected.

SLT

  • Glue tables generated by SLT Add-On were missing the description of fields. Also, the generated Queue fetchers were missing RFCDEST and USE_DEDUPLICATION parameters. The logic has been adjusted and all the objects are created properly now.

  • The stability of the extraction has been improved. Especially the extraction when the number of jobs was smaller than the number of tables that were being transferred.

  • The error handling was improved when adding an Extraction Process to Queue Scheduler.

Object Generator

  • Object generation failed with a short dump if non-existing storage was specified. This has been corrected and a proper error message with a description of the problem is shown.

Additional Features

  • The calculation of Glue statistics has been optimized for large numbers.

  • Glue Data Browser was not able to display the content of tables with the name starting with '/'.

  • In the new version, the verification of authorization checks has been added to the missing places. Now the access to both Glue and Storage Management screens and functions are fully controllable using our authorizations.

Known issues

Extractors

  • Listcube fetcher doesn’t support the extraction of data from an Infoset.

  • Parallel extraction is not available for ODP. Nor is the extraction ODQ/ODP based on CDS views.

Storage Management

  • Merge operation is not supported for Google BigQuery storage.