tiflow v6.5.1
Improvements
- Support batch
UPDATE
statements to improve TiCDC replication performance #8084 @amyangfei - Enable pull-based sink to optimize system throughput #8232 @hi-rustin
- Support storing redo logs to GCS-compatible or Azure-compatible object storage #7987 @CharlesCheung96
- Implement MQ sink and MySQL sink in the asynchronous mode to improve the sink throughput #5928 @amyangfei @CharlesCheung96
Bug fixes
-
Fix the issue that changefeed might get stuck in special scenarios such as when scaling in or scaling out TiKV or TiCDC nodes #8174 @hicqu
-
Fix the issue that precheck is not performed on the storage path of redo log #6335 @CharlesCheung96
-
Fix the issue of insufficient duration that redo log can tolerate for S3 storage failure #8089 @CharlesCheung96
-
Fix the issue that
transaction_atomicity
andprotocol
cannot be updated via the configuration file #7935 @CharlesCheung96 -
Fix the issue that the checkpoint cannot advance when TiCDC replicates an excessively large number of tables #8004 @overvenus
-
Fix the issue that applying redo log might cause OOM when the replication lag is excessively high #8085 @CharlesCheung96
-
Fix the issue that the performance degrades when redo log is enabled to write meta #8074 @CharlesCheung96
-
Fix a bug that the context deadline is exceeded when TiCDC replicates data without splitting large transactions #7982 @hi-rustin
-
Fix the issue that pausing a changefeed when PD is abnormal results in incorrect status #8330 @sdojjy
-
Fix the data inconsistency that occurs when replicating data to a TiDB or MySQL sink and when
CHARACTER SET
is specified on the column that has the non-null unique index without a primary key #8420 @asddongmen -
Fix the panic issue in table scheduling and blackhole sink #8024 #8142 @hicqu
-
Fix the issue that the binlog-schema delete command fails to execute #7373 @liumengya94
-
Fix the issue that the checkpoint does not advance when the last binlog is a skipped DDL #8175 @D3Hunter
-
Fix a bug that when the expression filters of both "update" and "non-update" types are specified in one table, all UPDATE statements are skipped #7831 @lance6716