| 2.0.22/2.1.8 | - Feature updates
- Supported CDC, which can directly and repeatedly backtrack/extract the binlogs in the custom log retention period. This solves the problem where the compute node loses local binlogs in scenarios such as HA. For more information on how to set the binlog retention period, see Setting Backup Retention Period.
- Optimized the pages purge rate to improve the database performance.
- Fixes
- Fixed the repeated crashes of the compute instance when the upper limit of the space was reached.
|
| 2.0.21/2.1.7 | - Feature updates
- Added the password dictionary parameter as described in Overview, and optimized HA's dependence on dictionary files.
- Supported completing purged binlogs in the kernel.
- Supported the built-in database proxy for the Serverless architecture to implement connection persistence and momentary disconnection prevention, and fixed the connection error reported during the first wakeup.
|
| 2.0.20/2.1.6 | - Fixes
- Fixed the issue where the liveness probe of the database proxy caused the kernel to output a large number of error logs, and blocked the printing of redundant logs.
- Fixed the issue where the uninitialized GCR LSN in the session LSN tracker might return a random value of the database proxy and thus cause the statement execution to time out.
- Fixed the issue where creating a temp table in a read-only instance might cause a deadlock.
|
| 2.0.19 | - Feature updates
- Supported logical backup for read-only instances.
- Supported database proxy. For more information on how to use this feature, see Database Proxy Overview.
- Supported parallel replication of binlogs at the table level.
- Supported setting the change buffer and merge modes.
- Supported `show detail processlist`.
- Fixes
- Fixed the official bug #22991924 related to the JSON character set.
- Merged the official bugfix #25865525 to solve the issue where `LOAD DATA INFILE` failed to read escape characters plus UTF8 characters.
- Merged the official bugfix #31529221 to fix the issue where the error `Incorrect key file` was reported upon ALTER TABLE failure.
- Merged several official bugfixes related to column generation and cascading deletion, including #33053297, #32124113, and #29127203.
- Fixed the official bug #31599938 where resetting the source caused a crash when binary logging was disabled in the replica.
- Fixed the issue where the startup time was prolonged due to the failure to add `dict op lock` to the main thread caused by large transactions or long-line transaction rollbacks during instance startup.
- Fixed the crash caused by TRX reuse after the failure to allocate the undo page.
- Fixed the crash caused by executing `alter table` on a partitioned table to migrate from the extended tablespace to the system tablespace.
- Fixed the crash caused by the startup before the truncate log was completely written.
- Fixed the crash caused by inserting data after `drop table partition force`.
- Fixed the possible crash caused by rollbacks after instant DDL.
- Fixed the issue where creating tables in the extended tablespace with `create temporary table like` failed.
- Fixed the OOM caused by the continuous increase of the cache during data writes to the full-text index table.
|
| 2.0.17 | - Feature updates
- Supported adding the binlog with the specified filename to an index file.
- Fixes
- Merged the official bugfix #25865525 to solve the issue where `LOAD DATA INFILE` failed to read escape characters plus UTF8 characters.
|
| 2.0.16 | - Performance optimizations
- Optimized `undo space truncate` to improve the speed of `undo truncate` on large-spec instances.
- Optimized the performance of large-scale queries on read-only instances.
- Fixes
- Fixed the issue where `backup lock` couldn't be locked due to the `lock table` statement.
- Fixed the issue where `table share` went wrong after thousands of columns were added through `instant add`.
- Fixed the replay error when the content of binlog contained escaped keywords.
- Fixed the issue where externally prepared XA transactions were not explicitly rolled back and thus blocked normal shutdown.
- Fixed the issue where the warning "tablespace -1 not found" was reported during read-only instance startup.
|
| 2.0.15 | - Feature updates
- Supported the extended table space: When a single table space exceeds the `innodb_ncdb_extend_space_threshold` configuration, a new table will be created in the extended table space.
- Added new JSON functions: JSON_MERGE_PRESERVE, JSON_MERGE_PATCH, JSON_PRETTY, JSON_STORAGE_SIZE, JSON_ARRAYAGG, JSON_OBJECTAGG.
- Optimized the table lock recovery process at system startup to shorten the startup time.
- Fixes
- Fixed the bugs for the `group by` performance issue in text columns and multiple issues related to virtual columns.
- Fixed the possible crash when large transaction rollback and shutdown operations were performed concurrently after instance startup.
- Fixed the issue where repeatedly failed IO retries of related pages caused instance exit after `undo space truncate` failed.
- Fixed the crash when statistics update accessed the snapshot cache for disabled read-only instances.
- Fixed the issue where the truncation log might be behind the `truncate` operation in `undo space truncate`.
- Fixed the bug where an error in ICP check for partitioned table scan resulted in slow query.
- Fixed the crash when the previous scan in a read-only instance encountered the partial replay of a split index log.
|
| 2.0.14 | - Feature updates
- Fixes
- Fixed the issue where the used space was not reclaimed when a temp table in a read-only instance was dropped.
- Fixed the issue where the process exited when a read-only instance read the old page version due to changes in storage routes.
- Fixed the issue of possible crash when `information_schema.metadata_locks` was queried.
- Fixed the concurrency error of forward scan and B-tree SMO in read-only instance.
- Fixed the issue where when multiple tables had complex foreign key dependencies and the foreign key attribute was `ON DELETE CASCADE`, the corresponding record in a child table might be deleted twice when a record was delete in its parent table with DELETE.
- Fixed the issue where the process exited due to operations such as CREATE USER in a read-only instance.
- Fixed the issue where SHOW VOLUME STATUS in a read-only instance might trigger an assertion failure.
- Fixed the issue where a read-only instance had abnormal query results and crashed when DDL operations were performed in partitioned tables frequently.
- Fixed the issue where the process crashed during historical version construction when an read-only instance scanned a purged undo log.
|
| 2.0.13 | - Feature updates
- Supported INSTANT ADD COLUMN. For more information, see Instant DDL Overview.
- Optimized the audit performance under high load and added the `lock_usleep_time` dynamic parameter.
- Fixes
- Fixed the issue where `dict_operation_lock` might cause deadlock during foreign key check.
- Fixed the issue where the process might exit when the ACL change log was replayed during read-only instance startup.
- Fixed the issue where data in source and replica instances was inconsistent after INSTANT ADD COLUMN and TRUNCATE TABLE.
- Fixed the log replay error occurring when data was inserted again after INSTANT ADD COLUMN and TRUNCATE TABLE.
- Fixed the issue where the process exited when a primary key containing a column added by INSTANT ADD COLUMN was created.
- Fixed the issue where the process exited during table structure query in a read-only instance when INSTANT COLUMN was used to create or rebuild a partition.
|
| 2.0.12 | - Feature updates
- Supported database audit. For more information on how to use it, see Enabling TDSQL-C for MySQL Audit.
- Supported purging page read-ahead to accelerate space reclaim.
- Supported real-time update of the size information of tables and indexes in the system view.
- Added a thread to accelerate recycle LSN and storage GC.
- Fixes
- Fixed official bugs in full-text index, including bugs #24938374, #21625016, #27082268, #27155294, #27326796, #27304661, #25289359, #29717909, and #30787535.
- Fixed official bugs where concurrent update might cause system crashes, including bugs #30950714, #31205266, and #25669686.
- Fixed the official bug #28104394 where uncommitted INSERT operations would affect the range scan created by an index and made it return an incorrect number of rows.
- Fixed the official bug #30488700 where an incorrect query execution plan of a derived table could result in a poor performance.
- Fixed the issue where the process exited when a read-only instance replayed statistics logs after the source (read-write) instance executed a DDL statement.
- Fixed the issue where RENAME TABLE was performed on a database that did not exist.
- Fixed the issue where a read-only instance might exit when OPTIMIZE TABLE was used for the source instance.
- Fixed the issue where transactions were inconsistent after a table with a full-text index was updated in a read-only instance.
- Fixed the deadlock occurring during SET OFFLINE MODE and SHOW VARIABLES operations.
|
| 2.0.11 | - Feature updates
- Optimized the binlog file index to accelerate binlog file scan.
- Optimized the shutdown process to make it faster.
- Optimized the instance memory to reduce the memory usage by structures such as buffer, ZIP, and hash.
- Optimized the DDL lock recovery process during read-only instance startup to accelerate replay.
- Optimized system table loading in read-only instance to accelerate startup.
- Optimized worker thread assignment during PURGE COORDINATOR to accelerate purge.
- Fixes
- Fixed the issue where the process crashed during OPEN TABLE due to incorrect index structure mapping.
- Fixed the issue where read-only instance replication was abnormal during XA transaction rollback.
- Fixed the issue where startup crashed when binlog replication was started in a read-only instance.
- Fixed the memory leak caused by FLUSH LOGS.
- Fixed the shutdown failure of `mysqladmin shutdown`.
- Fixed the issue where a read-only instance failed to replay logs when DROP COLUMN was performed on the source instance.
- Fixed the issue where data could still be input after space restriction was triggered when a BLOB was inserted.
- Fixed the issue of read-only instance replication availability that might be caused by DDL statements in big tables or slow log storage in the source instance.
- Fixed the issue where storage wasted small tables after `innodb_max_temp_data_file_size` was set for a temp table.
|
Yes
No
Was this page helpful?