Skip to main content

BladePipe 1.2.0

Release Date: January 15, 2026

Version: 1.2.0

Highlights​

  • Newly support KingbaseES V8/V9 as source.

New Connections​

  • KingbaseES V8/V9 > MySQL / StarRocks / Doris / SelectDB: full data migration, incremental data sync, data verification, data correction, schema migration, and DDL sync (add column / drop column / modify / rename / truncate).
  • KingbaseES V8/V9 > ClickHouse: full data migration, incremental data sync, data verification, schema migration, and DDL sync (add column / drop column / modify / rename / truncate).
  • OceanBase for MySQL > Oracle: full data migration, incremental data sync, data verification, data correction, schema migration, and DDL sync (add column / drop column / modify / rename / truncate).
  • Doris / SelectDB > Doris / SelectDB: full data migration, data verification, data correction, and schema migration.

New Features​

  • Support OceanBase for MySQL / TiDB / PolarDB MySQL / TDSQL-C MySQL and other MySQL-protocol-compatible databases as the console metadata database.
  • Support duplicate subscription detection when creating DataJobs or modifying existing DataJobs, making it easier for users to reduce the number of DataJobs.
  • Support MySQL source RENAME_COL type DDL, and support synchronization of this type of DDL to StarRocks (≥ 3.3.2).
  • Support skipping unsupported variable part events when parsing MySQL binlogs.
  • Support virtual columns for Hana > Doris and OceanBase > StarRocks.
  • Support SQL Server source heartbeat, resolving latency issues when there is no write on the source.
  • Support Doris / SelectDB > MySQL data verification and data correction.
  • Support DB2 source metadata retrieval, making it easier for users to view subscription relationships.
  • Support DB2 source performance parameter template whitelisting.
  • Support setting the automatic commit time after transaction timeout on the Dameng source. (Parameter: autoCommitTxTimeoutSec)
  • Support keeping a certain interval between the Dameng source consumption position and the latest Arch log. (Parameter: fallBackLsnStep)
  • Support displaying the latest commit delay on the Dameng source to quickly determine DataJob latency caused by uncommitted transactions.
  • When Doris is used as the target, schema migration supports configuring partition information, including: Automatic / manual RANGE partitions, LIST partitions.
  • Support ElasticSearch target completion type.
  • When creating DataJobs, table and field mapping rules support: CamelCase to snake_case, snake_case to CamelCase.
  • Support canceling test connection when creating DataJobs or adding DataSources.

Improvements​

  • Optimized MySQL source synchronization so that Online DDL uses DDL Convert logic by default, significantly improving compatibility of Online DDL synchronization.
  • Optimized filtering of invalid default datetime / timestamp values during MySQL source schema migration.
  • Optimized MySQL full migration for multiple schemas. Improved startup speed and reduced connection pool usage.
  • Optimized MySQL source invalid date / datetime / timestamp data judgment logic.
  • Optimized PostgreSQL full migration to support primary key–based pagination for tables with multiple primary keys (previously snapshot read).
  • Optimized the time zone offset issue in SQL Server latency display.
  • Optimized SQL Server source behavior for versions 2014 and below, setting SnapshotRead to true by default, and using snapshot read for scanning.
  • Optimized metadata definition retrieval during SQL Server / Oracle DataJob startup, improving retrieval speed by 10x.
  • Optimized MySQL source initialization in non-GTID mode, no longer storing GTID positions to prevent failures caused by excessively long position information.
  • Optimized the default value of maxInSizePerQuery for Doris / StarRocks targets to 1024.
  • Optimized DB2 source full scan using a unified architecture, consistent with other data sources.
  • Optimized DB2 source behavior so that when snapshot read is enabled, position range retrieval is no longer performed, resolving timeout issues.
  • Optimized DB2 source metadata reading using pagination to support retrieving metadata for a large number of tables.
  • Optimized permission pre-checks for Hana source incremental DataTask, reducing the scope to the table level.
  • Optimized Dameng source log printing, adding query time statistics for V$LOGMNR_CONTENTS.
  • Optimized the subscription modification process by adding a secondary confirmation dialog.
  • Optimized redirection to the DataJob list when entering an incorrect or non-existent route.
  • Optimized backend behavior so that when multiple APIs return errors simultaneously, all error messages are collected and displayed in a single pageable dialog.
  • Optimized display of pending creation status in database mapping tree view.
  • Optimized frontend single-request timeout, extending it to 1 hour.
  • Optimized display of connection information for Console and Worker automated O&M operations.
  • Optimized batch blacklist behavior. If no new tables are added, the effective scope of newly added tables is disabled.
  • Optimized table selection performance when creating DataJobs with a large number of tables (20,000+), improving browser stability during “select all”.

Bug Fixes​

  • Fixed connection issues when adding an Oracle data source with the username configured as username as role (e.g. sys as sysdba).
  • Fixed an issue where duplicate rows could not be inserted on the target during incremental synchronization for Oracle tables without constraints.
  • Fixed an issue where Oracle index columns were incorrectly retrieved due to missing owner restrictions, causing columns with the same index name in other schemas to be retrieved.
  • Fixed an issue where parsing BC timestamp with/without time zone data failed when strong typing was enabled for PostgreSQL incremental synchronization.
  • Fixed an issue where incremental synchronization DataJobs for OceanBase / OceanBase for Oracle entered infinite reconnection after the LogProxy service process exited. (Added parameter: logProxyMaxReconnectTimes)
  • Fixed an issue where failure to parse DDL on OceanBase for Oracle caused the log listening client to stop while the DataJob process remained running.
  • Fixed an issue where the main DataJob could not be started after DB2 source subtask merging.
  • Fixed an issue where modifying the DB2 source position did not take effect.
  • Fixed missing script display when adding virtual columns with expression-based values.
  • Fixed an issue where unsupported target columns were selected during data processing after switching the target when creating similar DataJobs.
  • Fixed an issue where parameters were missing when calling the interface to cancel Worker automatic upgrades.
  • Fixed metadata mapping errors when synchronizing table names containing uppercase letters under MySQL source case-insensitive configuration.
  • Fixed an issue where UK identifiers were not displayed for Doris source table information.
  • Fixed an issue where schema migration partition information was lost when multiple spaces existed between PARTITION BY in StarRocks partition configuration.
  • Fixed an issue where abnormal state machine states during SQL Server DataJob creation caused positions to fail to initialize.
  • Fixed an issue where Worker automatic deployment used high-privilege accounts, resulting in incorrect deployment directory permissions.
  • Fixed an error when retrieving DataJob lists via paginated API calls with an empty sort order.
  • Fixed an issue where DDL add-column execution failed for ElasticSearch targets.
  • Fixed an issue where editing a single database was ineffective in database mapping tree view.
  • Fixed an issue where console O&M action forms could be submitted before validation was completed.
  • Fixed rendering issues when position information in DataJob details was excessively long.
  • Fixed an issue where error interfaces could be clicked repeatedly in a short period, triggering a large number of requests.
  • Fixed an issue where dialogs were not closed when navigating back in the browser.
  • Fixed a 404 issue with the partition expression API.
  • Fixed an issue where unsupported target tables were selected during table selection after switching targets when creating similar DataJobs.
  • Fixed incomplete database mapping validation when creating similar DataJobs, which previously allowed proceeding to the next step without full validation.
  • Fixed an issue where target table information was not refreshed after switching target schemas during DataJob creation or similar DataJob creation.
  • Fixed an issue where message-based source DataJobs showed target tables in pending creation status after clicking “select all”.
  • Fixed console errors when modifying subscriptions for message-based sources to PostgreSQL / SQL Server.
  • Fixed missing dropdown options for target mapping when adding Topics during subscription modification from message-based sources to Oracle.
  • Fixed an issue where PostgreSQL full migration caused index invalidation due to implicit type conversion of primary key fields (smallint / integer / numeric / bigint).
  • Fixed an issue where UPDATE events could not be synchronized when strong typing was enabled for PostgreSQL incremental synchronization.
  • Fixed inconsistency between PostgreSQL source and target when strong typing was enabled and writing BC Timestamp with time zone.