Release 4.3.0 + 4.4.0

Author: heywalter

docs/data-replication/incremental-check.md

@@ -2,10 +2,6 @@
 
 Incremental data validation is a real-time mechanism designed to enhance data accuracy and consistency. It periodically samples and compares newly inserted or updated records between the source and target systems to detect and automatically correct inconsistencies.
 
-## Prerequisites
-
-Tables selected for incremental validation must have a primary key or a unique index.
-
 ## Background
 
 As real-time data synchronization and integration become core demands across modern enterprises, ensuring consistent data flow between heterogeneous systems has grown increasingly critical. However, real-time sync is often prone to issues such as network latency, system errors, and failed writes, which can result in inconsistencies between source and target databases—ultimately impacting business decisions and operational efficiency.

docs/images/api_versioning_demo.png

@@ -58,6 +58,8 @@ Click the **Result** link for any validation task to view its results in depth.
 - Use **One-Click Repair** to automatically align data between source and target.
 - Download detailed mismatch reports for further analysis.
 
+Additionally, you can **generate repair SQL** for in-depth analysis or execute it manually in the database to fix the issue.
+
 ![View validation Results](../../images/check_data_result_en.png)
 
 For **Full Table Value Validation** or **Key Field Value Validation**, you can also click **Diff Validation** in the top-right corner to re-run validation only on the previously identified differences, confirming whether they have been resolved.

docs/images/check_data_result_en.png

@@ -14,6 +14,29 @@ import TabItem from '@theme/TabItem';
 <TabItem value="Version 4.x" default>
 ```
 
+## 4.4.0
+
+### New Features
+
+- [Incremental data validation](data-replication/incremental-check.md) now supports tables without primary keys, along with manual repair. This expands validation coverage and helps ensure data consistency.
+- Added [alert notifications](system-admin/other-settings/notification.md#alert-settings) during task retries to improve visibility and responsiveness in failure scenarios.
+- Introduced "Node Data Preview" in the task editor, allowing you to view sample fields instantly while configuring tasks—speeding up design and troubleshooting.
+
+### Enhancements
+
+- Improved data source monitoring: Added connection status and performance metrics for Sybase and PostgreSQL, enhancing system observability.
+
+## 4.3.0
+
+### New Features
+
+- [Data validation](operational-data-hub/fdm-layer/validate-data-quality.md) now supports exporting repair SQL for inconsistent results, with compatibility for PostgreSQL, Sybase, and **SQL Server**. This makes offline auditing and rollback easier for ops teams.
+- JS Processor Node now supports the `unset` operation, enabling targeted field removal in MongoDB. Ideal for cleanup and payload optimization.
+
+### Enhancements
+
+- Refined logic for marking full sync tasks as completed: now based on whether all writes to the target are finished—more aligned with real business completion.
+
 ## 4.2.0
 
 ### New Features