Here are highlights of the significant changes in Syniti Replicate v10.7.0.4.
New Features
Support for Databricks Unity Catalog
Syniti Replicate v10.7.0.4 introduces support for Databricks with Unity Catalog enabled as a target system, allowing for seamless integration with your analytics and machine learning workflows.
Customers can utilize Syniti Replicate v10.7.0.4 for Snapshots (Refresh) using the Bulk Insert feature. With this feature, users can replicate large volumes of data to Databricks, reducing data loading time and minimizing downtime. For customers that need high-precision data, Syniti introduces the Change Data Capture (CDC) technology using the Bulk Mirroring feature. The latest CDC technology ensures that only the changes made to your source data are replicated to Databricks, minimizing latency and ensuring data consistency across your systems.
Refer to Databricks for more information.
Support for SAP Systems via RFC
Syniti enhances our current support for the SAP’s ECC and S/4HANA systems via Remote Function Call (RFC), introducing a new Syniti Data Provider driver, including Streaming and CDC capabilities. Designed with precision and versatility in mind, this cutting-edge driver represents a significant advancement in SAP connectivity, offering unparalleled efficiency and reliability for your business operations.
Refer to SAP ECC via Syniti RFC and SAP S/4HANA via Syniti RFC for more information.
Bug Fixes
Replication Agent
We fixed an issue when mirroring to Azure Data Lake Storage (ADLS) Gen2 where the MIR files were available in Syniti’s temporary directory and were not uploaded to ADLS Gen 2.
We fixed an issue when refreshing big tables with Partitions in ADLS Gen2 that displayed the following error: The process cannot access the file '...._ref_269.tmp' because it is being used by another process.
We fixed an issue when using the DELETE JOIN query during Bulk Mirroring in MySQL v8.
Bulk Mirroring now functions correctly when using MySQL GCloud v5.7 as a target.
We fixed an issue where the Bulk Insert Refresh failed due to the following error: LOAD DATA LOCAL INFILE command is not supported in the prepared statement protocol yet for MySQL databases.
We fixed an issue when creating a replication to the MySQL target database with the DATETIME(3) field being mirrored, where the corresponding staging table is generated without the precision of the DATETIME field.
We fixed an issue where the Tab characters were not processed in Snowflake while performing Bulk Insert.
We fixed an issue where Syniti Replicate displayed an error when uploading .CSV files to Google BigQuery that contained data with Quote (") symbols.
We fixed an issue where the Refresh partitions displayed the Null Reference exception when some partitions were disabled.
We fixed the following Oracle trigger error when replicating in the Mirroring mode from Oracle: missing SET keyword.
Replication Agent now functions correctly without displaying an error when you create mirroring replications using Triggers from HANA and Oracle with the below two options:
Uncommitted Transactional Recovery
Continue and Process when Committed
We fixed a logical issue with the Uncommitted Transactional Recovery option using Triggers with HANA.
We updated the READDB2UDB_LOGREC03 stored procedure to resolve the Missing Records issue during replication from the IBM Db2 LUW source.
We fixed the following parsing error when performing mirroring replications from IBM DB2 as a source: Input string was not in a correct format.
Management Center
We fixed an issue where the numeric/decimal data type conversion functioned incorrectly when using SQL Server as a Source and MySQL as a Target.
We resolved a regression issue affecting connection testing from the Management Center to connection with Azure Data Lake.
We fixed a script error issue when creating target tables with column names starting with the Hash (#) character in the Create Target Table Wizard.
In Verifier Options > Reconcile, the following checkboxes are now selected by default for the Operation Types To Reconcile field. This update enables you to resolve any data differences between source and target tables using the default configuration.
We fixed an issue where the Export Replications Wizard displayed an error on the Processing page.
We fixed an issue where the Group Name of the Collection Group type was not processed in the replication script.
We fixed an issue while removing the staging tables using the Yes All feature when switching the replication modes for multiple replications in a group.
Note
Other bug fixes and performance improvements can also be observed in Syniti Replicate v10.7.0.4.