(Glue 2402) Release Notes 2402

Owned by SNP
Feb 22, 2024
3 min read

Please note that as of release 2402, Trigger and Queue watchdogs, and Extraction Scheduler are no longer supported.

We highly recommend using Glue Scheduler instead.

You can use the following guidelines for the migration of existing configurations of watchdogs and Extraction Scheduler to the new Glue scheduler.

Snowflake drivers update

To ensure smooth operation when utilizing both the Snowflake connection and Snowpipe Streaming connection simultaneously, it is advisable to employ two separate directories for drivers. One is designated for Snowflake and another for Snowflake Streaming. This segregation helps mitigate inconsistencies that may arise between Snowflake driver versions and the Snowflake Ingest SDK.

At present, Snowpipe Streaming requires JDBC driver version 3.13.13. In contrast, the standard Snowflake connection is compatible with the latest versions of the JDBC driver. Therefore, maintaining distinct directories for drivers facilitates the proper compatibility and functionality of both connection types.

Snowflake (Snowpipe Streaming)

In cases where the user wants to utilize both Snowflake connection and Snowpipe Streaming connections simultaneously, it is required to set up separate Java Connector instances for Streaming to avoid inconsistencies.

New Features

SAP Datasphere

You can now use SNP Glue to transfer data to SAP Datasphere and thus leverage the capabilities of the next generation of SAP Data Warehouse Cloud.

AVRO file support

With the latest release, Glue supports the AVRO file format. This file format can be now used in 2 scenarios:

  1. Extraction of data to file storage. In this case, the data can be stored in AVRO files (previously CSV and parquet file formats were supported).

  2. Ingestion of data to Kafka using AVRO files.

CDS Fetcher

A new fetcher type has been introduced that enables you to extract data from CDS views/entities.

Improvements

Extractors

  • ListCube fetcher now provides an option to disable navigation attributes which is useful when fetching data from specific info providers.

  • You can now replicate data from SAP tables using the automated parallelization mechanism from multiple clients. In the previous release, only replication from a single client was supported.

Storage Management

  • Snowflake Snowpipe streaming connector has been improved so that it can handle large data loads stably and consistently.

  • For connection to AWS Redshift, you can now use IAM authentication.

  • The performance of Azure SQL storage has been improved by providing an option to create tables with clustered or nonclustered PKs and enabling bulk insert customizations.

Fixes

Extractors

  • When extracting data from an ODP queue using an ODP fetcher, technical fields from an ODP were blank. This has been adjusted and the data is now available.

  • The automated mass execution didn’t work properly when the same field was used for both filtering and splitting.

  • The extraction with trigger delta was failing if there was a selection defined that contained non-key fields.

General

  • The performance of the Glue Cleanup watchdog has been improved.

  • In the Glue cockpit, the Queue size value hasn’t been updated properly. This has been fixed and the correct number is now displayed.

  • Glue TMS didn’t transport correctly Google BigQuery Cortex-specific settings. This has been corrected and all the settings are now correctly transported.

  • Folders in the Glue Object Navigator (GL80) have been sorted alphabetically.

Storage Management

  • When configuring an AWS S3 connection, the field length for the KMS key was insufficient. The field length has been increased and now you can set up the connection without any problems.

  • Since the Glue release 2311, the ADLS Gen2 connector provides an option to choose whether you want to use the SAP ICM component or SNP JCO instead for the HTTP communication. When SNP JCO was picked, the OAuth 2.0 authentication didn’t work.