PolarDB:PolarDB for MySQL 8.0.1 release notes

Description

• Added a feature in Outline to support table partitioning scenarios. A single Outline can match all statements with the same template, applicable to different partitioned tables.

• Added the ability to dynamically adjust the selection result of Order Index during execution based on actual conditions.

• Added the parallel parsing of Redo feature for physical replication to improve physical replication performance.

• Added the variable ignore_use_force_index. When enabled, it ignores all use or force index in statements and Outlines (but does not affect Index() in Hints).

• Added string conversion functions convert_sys_to_filename() and convert_filename_to_sys() for converting between system_charset_info encoding and my_charset_filename encoding.

• Added support for setting transparent data encryption (TDE) for OSS cold data.

• Added support for enabling asymmetric MTT buffers for RW and RO nodes. Supports enabling MTT optimization in conjunction between RO and RW nodes.

• The IMCI i_s.imci_index_stats table now includes the PRUNED_PERCENT field, which indicates the Data Skipping ratio for each IMCI.

• Added a feature to the SHOW ProcessList result to display whether a session is waiting for thread pool scheduling.

• Supports rebuilding an IMCI by modifying the pack_shift in the COMMENT, without needing to delete it beforehand.

Description

• Optimized the dynamic Buffer Pool scale-in process to reduce lock conflicts.

• Optimized SQL Trace so that when Call procedure is used, the SQL and execution plan can be correctly recorded.

• Optimized the memory overhead of columnstore wide tables.

• Optimized the startup speed of columnstore read-only nodes. Supports independent startup of the columnstore engine from the columnstore snapshot offset.

• Optimized the thread startup and destruction code for logical pre-reading and simplified the memory check logic.

• Optimized the long startup time of column store nodes in scenarios with a massive number of tables.

• Optimized the performance and memory usage of IMCI for large transactions.

• Optimized the performance of the materialized HASH KEY expression in IMCI.

• Optimized the Serverless process for column store nodes. Added support for column store nodes to be added to the Proxy node to handle service traffic only after the index is ready.

• Optimized the statistics sampling mechanism of IMCI, increasing the sampling frequency and the proportion of sampled data.

  Optimized the sampling logic for low-cardinality columns to avoid situations where inaccurate statistics lead to queries not being fully optimized.

• Optimized the Simulated-AIO asynchronous performance blocking module.

Description

• Fixed an issue where executing a DDL on a table containing a view could result in an error or have no effect after an HA (node switchover).

• Fixed an issue where an RO node might remain unavailable after reconnecting to an RW node.

• Fixed a process crash issue caused by the CAST(REGEXP_SUBSTR(...)) AS DECIMAL statement.

• Fixed an issue where REGEXP_SUBSTR might return an incorrect NULL flag.

• Fixed an issue where the table creation statement for the dlm_policies system table was recorded in the BINLOG, causing downstream synchronization to fail.

• Fixed an issue where unique indexes were not fully utilized for acceleration in some queries.

• Fixed an issue where using the ALTER TABLE DROP statement to delete multiple full-text indexes (FTS) simultaneously could leave residual metadata.

• Fixed an unnecessary index rebuild issue caused by specifying a DEFAULT value when extending the length of a VARCHAR column.

• Fixed compatibility, expansion, and contraction anomalies, and other issues between IMCI Serverless and internal memory elasticity.

• Fixed an issue where full-text index retrieval would get stuck on a read-only node after a DDL operation.

• Fixed an issue in adaptive execution where certain types of queries were not routed correctly.

• Fixed an availability issue with the IMCI caused by the lack of support for online rebuilding.

• Fixed a crash issue caused by not fully adapting to the DDL global lock when batch adding columnstore indexes in a multi-write database/table cluster.

• Fixed a query crash issue in JSON_TABLE scenarios with NULL Fields.

• Fixed a conflict issue between multi-master Buffer Pool Resize and transaction access to blocks, and optimized the performance of multi-master Buffer Pool RDMA register or unregister.

• Fixed an issue where Digest CCL did not support the PS (Prepared Statements) protocol.

• Fixed an issue where an RO node could crash during the HA phase after the RO No Sync feature was enabled.

• Fixed an issue where a transaction rollback required a pessimistic lock to delete an index when using a Spatial Index, leading to an infinite loop and prolonged lock holding.

• Fixed a rare issue where the upgrade task from a historical version to 8.0.1.1.45.1 could be interrupted.

• Fixed a rare issue where accessing string data types during columnstore execution could cause a query error due to accessing invalid data.

• Fixed an issue where IMCI would fall back to row store when processing an expression containing multiple subqueries with SUM.

• Fixed a circular wait issue with asynchronous column index rebuilding when using Hybrid Plan to accelerate wide table queries was enabled in a one-writer, multiple-reader IMCI setup.

• Fixed an issue where Redo was written incorrectly for some pages if the innodb_bulk_load_page_grained_redo_enable parameter was modified within a certain time window during parallel DDL execution.

• Fixed a potential deadlock issue in BLINK for batch data deletion scenarios in tables containing LOB fields.

• Fixed a crash during the migrate process caused by a concurrency issue between high-frequency INSTANT DDL operations and background Compaction tasks.

• Fixed an issue where GROUP BY could cause a temporary table to report a table is full error.

• Fixed an issue where the index join in IMCI could lead to unstable results when using a constant as a join condition.

Description

• Fixed an issue where the table Autoinc value would roll back after executing an INSERT ... ON DUPLICATE KEY UPDATE statement following an RO node to RW node switchover.

• Fixed compatibility, expansion, and contraction anomalies, and other issues between IMCI Serverless and internal memory elasticity.

• Fixed a bug that caused a crash when the number of tables exceeded 61 during query transformation. Query rewrite is now skipped in this scenario.

Description

• Fixed an issue where full-text index retrieval would get stuck on a read-only node after a DDL operation.

• Fixed an issue where the backward index scan feature was incompatible with optimizations such as ICP. A new polar_optimizer_switch switch has been added for control.

Description

Added the variable ignore_use_force_index. When enabled, it ignores all use/force index in statements and Outlines (but does not affect index() in hints).

Description

• The INFORMATION_SCHEMA.IMCI_INDEX_STATS table in IMCI now includes the PRUNED_PERCENT field, which indicates the data skipping ratio for each columnstore index.

• Optimized SQL trace so that when the SQL trace type is set to SLOW LOG, SQL trace will also apply to statements specified by DEMAND.

• Optimized RW and RO to support asymmetric MTT buffers. Supports enabling MTT in conjunction between RO and RW nodes.

• Optimized the Serverless process for column store nodes.

  Allows new column store nodes to serve traffic only after their indexes are ready, reducing slow queries caused by SQL statements being routed to new nodes and running row-store plans during this period.

• Optimized the performance overhead of saving partition information during SQL execution in columnstore index (IMCI), improving partitioned table query performance.

Description

• Fixed a cluster crash issue during the Undo initialization phase when Force recovery was enabled.

• Fixed a circular wait issue with asynchronous column index rebuilding when Hybrid Plan was enabled in a one-writer, multiple-reader IMCI setup.

• Fixed an error when re-importing the same data into a UNIQUE CHECK IGNORE partitioned table using LOAD DATA with REPLACE.

• Fixed an issue in a three-zone deployment where recent_written could introduce dirty data if the CRC calculation value of the incoming data was exactly 0.

• Fixed an issue where an RO node could get stuck during the HA (node switchover) phase, preventing excessively long user HA times.

• Fixed B-tree corruption when adding an auto-increment column and creating a descending primary key index on that column.

• Fixed an issue where IMCI would fall back to row store when processing an expression containing multiple subqueries with SUM.

• Fixed a rare memory residue issue when an error was encountered during parallel primary key rebuilding.

• Fixed a potential crash issue when processing LOB fields in a GROUP BY query while using the Temptable engine.

• Fixed an issue where parallel query results could be incorrect when the predicate condition contained a subquery.

• Fixed a potential query crash issue in JSON_TABLE scenarios with a NULL FieId.

• Fixed an NCI encoding anomaly for EQUAL_PACK_LENGTH type DDL operations when modifying a PRIMARY KEY field.

• Fixed a conflict between Proxy advanced transaction splitting and Fast Query Cache that affected the result set.

• Fixed an issue where using a JSON_ARRAY expression in a subquery could lead to incorrect results when IMCI was enabled.

• Fixed a process interruption issue in the new process for database and table restoration after multiple rounds of restoration on the same table.

• Fixed a memory leak issue during the dynamic enabling process of the RDMA log transfer feature.

• Fixed an availability issue with the IMCI caused by the lack of support for online rebuilding.

• Fixed an issue where asynchronous DDL could not be used when adding a IMCI by modifying the COMMENT on a table with a full-text index.

• Fixed an issue where transactions might not be killed in high-concurrency scenarios.

Description

• Fixed a rare issue where the upgrade task from a historical version to 8.0.1.1.45.1 could be interrupted.

• Fixed a rare issue where accessing string data types during a columnstore execution query could cause a query error due to accessing invalid data.

Description

Optimized the memory overhead of IMCI columnstore wide tables.

Description

• Added a display to show whether a session is waiting for thread pool scheduling.

• Added a join reorder strategy based on a greedy algorithm in IMCI.

• Added a stored procedure and extended SQL command to batch add IMCI to an entire database.

• Added a major engine version upgrade feature from a PolarDB 5.7 backup set to improve one-click migration efficiency.

• The Hybrid Plan in IMCI now supports partitioned tables, virtual columns, and columns of ENUM, SET, and GIS types.

Description

• Optimized PolarDB IO Threads to be dynamically adjustable without a restart.

• Optimized the automatic elasticity of IMCI single-machine replay to reduce replay latency and out-of-memory (OOM) issues caused by high replay pressure.

• Optimized the read performance of Date type fields in IMCI.

• Optimized the IMCI statistics sampling mechanism, increasing the sampling frequency and the proportion of sampled data.

  Optimized the sampling logic for low-cardinality columns to avoid situations where inaccurate statistics lead to queries not being fully optimized.

Description

• Fixed a node crash issue caused by InnoDB full-text indexes in a multi-master cluster.

• Fixed an issue where the optimizer significantly misestimated the selectivity when a JOIN condition in IMCI included a type conversion.

  Fixed an issue where the Hybrid Plan could not be correctly selected in IMCI due to an overestimation of the output cardinality of a JOIN.

• Fixed an issue where JSON_LENGTH returned incorrect results when wildcards were present.

• Fixed an issue where Redo was written incorrectly for some pages if the innodb_bulk_load_page_grained_redo_enable parameter was modified within a certain time window during parallel DDL execution.

• Fixed an issue where the Autoinc value of a partitioned table would roll back after a read-only node was switched to a read/write node.

• Fixed an issue where removing an RO node could cause monitoring connections to be blocked for a period when innodb_polar_log_rdma_transfer was enabled.

• Fixed an issue where the index join in IMCI could produce unstable results when using a constant as a JOIN condition.

• Fixed an issue where a temporary table using a group by statement could cause temptable to report a table is full error even when the disk was not full.

• Fixed an issue where user threads waited too long for a Free Page when the Buffer Pool on an RO node was very full.

• Fixed a potential deadlock issue in blink for batch data deletion scenarios in tables containing lob fields.

• Fixed an issue where applying JOIN elimination optimization in IMCI could lead to unstable results.

• Fixed an issue where the Autoinc value would restart from 1 on the next write operation after all data in a table was deleted.

• Fixed a drop database failure issue when a database contained tens of thousands of tables.

• Fixed an issue where useless subqueries were not deleted during the view merging process.

• Fixed an issue where a query could report an out-of-memory error when executed in IMCI if it had too many UNION ALL clauses.

• Fixed an issue where a desc select statement could not correctly call an outline.

• Fixed an issue where table_comments in information_schema.tables displayed xxx is not base table.

• Fixed an issue where IMCI's implicit conversion of date types was incompatible with MySQL's behavior, leading to unexpected results.

• Fixed a query result inconsistency issue caused by a mismatch in the default pruner type when adding a new partition to a partitioned table created in an older version after an upgrade.

• Fixed a floating-point precision issue in JSON.

Description

• Supports deriving corresponding views from join conditions.

• Added a late materialization version of UnionAll for IMCI.

Description

• Optimized the time taken for query optimization on simple queries (single-table queries).

• Multi-master clusters now support fine-grained table-level locks.

• Optimized the performance of decimal sum/avg in IMCI.

• Optimized performance in hot spot update scenarios.

• Optimized the performance of hash match in IMCI.

• Optimized the memory for metadata used in IMCI queries.

• IMCI now supports filter pushdown to outer join.

• Adjusted the PolarFS transaction processing thread priority to improve PolarDB DDL performance.

Description

• Fixed an issue where import table failed due to a blink dangling page.

• Fixed an issue where the DB would hang during a manually triggered checkpoint in various processes.

• Fixed an issue where frequently toggling RO No Sync dynamically could cause markers to be lost.

• Fixed an issue where a Standby node in a multi-master cluster could not start due to an abnormal rollback segment.

• Fixed an issue where an InnoDB table with a function index could not be archived.

• Fixed an issue where Last_query_cost could become negative in a few optimization scenarios.

• Fixed an issue where join elimination could not be performed when a function contained an aggregate function.

• Fixed an issue where the IP address in audit logs was not updated in a timely manner when connections were reused under a session-level connection pool.

• Fixed issues related to acquiring and releasing global table locks in a multi-master cluster.

Description

• Added support for Binlog in transaction resumable upload.

• Added the parameter thread_pool_reset_waiting_state_before_queuing.

Description

• Columnstore index IMCI added an Index Join based on Hybrid Plan, which supports using InnoDB indexes for Join operations.

• Optimized the Page read performance of read-only nodes on PFS.

• Optimized the memory for metadata used in IMCI queries.

Description

• Fixed an issue where setting optimizer_switch and parallel_query_switch values would report an Error instead of a Warning for certain incorrect values.

• Fixed an issue where the optimizer_switch value would not take effect if a comma was added at the end.

• Fixed an issue where index selection changed after adding a limit operation to an implicit aggregate query.

• Fixed an issue where the undo log grew continuously after a follower was rebuilt in a 3AZ deployment.

• Fixed an issue where Persist RO could cause the RO to read old data.

• Fixed an issue of high replication delay on read-only nodes in a multi-master cluster.

• Fixed an issue where the optimizer could not select the optimal Join Order due to an error in cardinality estimation.

• Fixed a deadlock issue between threads that could occur during B-tree structure adjustment caused by record deletion.

• Fixed an issue where archiving an archived table with a descending index failed.

• Fixed an issue where no data was returned when using Limit Offset pushdown optimization combined with a secondary index equivalent condition and a primary key In List condition to get reverse-ordered paged output of the primary key.

• Fixed an issue of excessive permissions during DLM policy execution.

• Fixed an issue where executing a Create Table Select statement in a multi-master cluster affected the availability of other nodes.

• Fixed an issue where system pages could be corrupted when executing DDL statements on multiple nodes simultaneously in a multi-master cluster.

Description

• Added predicate pushdown functionality, including pushing predicates from HAVING to WHERE, and from HAVING to derived and subquery.

• Added a hybrid row-column optimizer and cost-based row-column routing capability.

• Supports viewing the invocation of Statement Outlines in the EXPLAIN result.

• Supports using a prefix murmur hash bloom filter in IMCI.

• Supports using the LEFT JOIN elimination feature in UPDATE statements.

• Supports using the instant column addition feature on tables with IMCI created.

• Added returned row count information to the Fast Query Cache. Queries that hit the cache can now display the number of returned rows in the audit log.

Description

• Introduced an execution feedback mechanism in row-column routing to collect real execution information and feed it back to the optimizer.

• Disabled index merge intersection by default for DML statements to improve performance and prevent deadlocks.

• Reduced the metadata overhead of memory allocation and memory statistics for IMCI to save memory.

• IMCI supports using BatchProbe to improve the performance of JOIN operations.

• Optimized the memory for metadata, file modules, and small RAM resident objects in IMCI to reduce resident memory usage.

• Optimized the error message for importing cold data from OSS back to InnoDB.

• When monitoring database performance, the impact of access by O&M engineers on the slow_queries status metric is ignored.

Description

• Fixed an issue where the OSS path mapping of a table in the recycle bin was incorrect after restarting the cluster when an OSS IBD table was recycled with the RECYCLE BIN feature enabled.

• Fixed an issue where a read-only (RO) node with excessive clock drift could preempt the read/write (RW) node's permissions when starting the NewHA vector voting disk.

• Fixed an issue where the dbms_oss.delete_table_file built-in stored procedure could not delete cold data in CSV format.

• Fixed an issue where the internal counter of the Thread Pool could become incorrect if an ACL lock wait was encountered during the asynchronous authentication process of the Thread Pool.

• Fixed an issue where the Blob space of the original record could not be reclaimed when performing an UPDATE operation on a Blob types of column.

• Fixed a data loss issue when repeatedly archiving an OSS foreign table.

• Fixed an issue where the optimizer selected an ordering index (sorting index), but the sorting operation was not omitted during actual execution.

• Fixed a node crash issue when creating an OSS foreign table because the oss_prefix parameter was not configured in the OSS Server.

• Fixed an issue where accessing some tables in the information_schema database under a multi-master cluster could retrieve abnormal data.

• Fixed an issue where performing a multiplication operation on Decimal type of data would return an assert message when at least one multiplier was 0 and the scale (decimal places) of the result exceeded 16.

• Fixed an issue where updating statistics on a partitioned table with an IMCI could result in inaccurate statistics because only some partitions were sampled.

• Fixed an issue where index equivalent access did not trigger adaptive row-column routing.

• Fixed an issue of excessive Autoinc init log printing.

• Fixed a potential error issue with OSS foreign tables in Order By Desc scenarios.

• Fixed a rare node crash issue when executing the TRUNCATE PARTITION command on a partitioned table with an auto-increment column.

• Fixed a data corruption issue caused by different index orders when executing an EXCHANGE PARTITION operation.

• Fixed an issue where the Buffer Pool's automatic cancellation of a preceding scale-in operation could cause a node crash under high concurrency.

Description

Added a Left Join feature based on TopK pushdown in columnstore index.

Description

• Supports granting users global dynamic permissions to execute the SHOW_ROUTINE, FLUSH_USER_RESOURCES, FLUSH_TABLES, FLUSH_STATUS, and FLUSH_OPTIMIZER_COSTS commands.

• Optimized the enabling process for global consistency high-performance mode. When the innodb_polar_scc parameter is set to ON on an RO node, the cluster automatically enables dependent features for high-performance global consistency, such as RDMA and PolarTrans, reducing the complexity of the control process.

• Supports using Variable-Setting Hint syntax in global consistency high-performance mode.

• Supports setting the polar_oss_ddl_shared parameter to ON to access table data that is being archived as cold data.

• Disabled the partial update feature for Blob fields.

• Supports using the database and table restoration feature for tables with an IMCI.

• Supports using the most common value statistics collection feature in IMCI to improve the accuracy of records per key estimation in data skew scenarios.

• Optimized the IMCI memory scheduling feature to improve memory utilization.

• Optimized the nonblocking DDL feature to support more DDL statements.

• Supports using the IMCI and global consistency (high-performance mode) features simultaneously.

• Supports using the CHECK TABLE command to check a specific partition.

Description

• Fixed an issue where a TRUNCATE statement could be executed on any node in a multi-master cluster.

• Fixed a node crash issue caused by accessing an invalid Blob reference when pre-reading Blob type data.

• Fixed a potential failure when exporting a local table to the OSS engine in parallel.

• Fixed an issue where the crash recovery logic could not run normally after an abnormal cluster restart because the LSN offset of the checkpoint did not meet expectations during a checkpoint operation.

• Fixed an issue where Autoinc would roll back and require a node restart to recover.

• Fixed a node crash issue that could occur when a CTE (Common Table Expression) referenced a TempTable engine-based temporary table multiple times and used an index scan on the temporary table.

• Fixed a memory out-of-bounds issue when IMCI processed extra-long string types.

• Fixed a probabilistic node crash issue when using the IMCI feature with the pre-read feature enabled simultaneously.

• Fixed an issue where the XEngine's compaction background task affected foreground operations and thus slow query performance due to overly strict concurrency control.

• Fixed an issue where using a BIGINT types of column in a Join operation could lead to inaccurate cardinality estimation and prevent the use of an IMCI if the column contained very large values.

• Fixed a deadlock issue in DDL read-ahead in parallel DDL scenarios.

• Fixed a memory leak issue that occurred during a Reload operation for features such as Statement Outline and Concurrency Control.

• Fixed an issue where BIT fields were displayed incorrectly after a Group By operation.

• Fixed an incompatibility issue between TDE and RDMA memory when forking to create a child process.

• Fixed an issue where duplicate groups appeared in the result set when a parallel query pushed GROUP BY down to a Worker for parallel execution in the presence of a partitioned table.

• Fixed an issue where executing an EXCHANGE PARTITION operation could cause index corruption and prevent access to table data due to different index orders.

• Fixed issues with partitioned tables whose primary key did not include the partition key during search, DML operations, and Binlog recording by the query optimizer.

Description

• Fixed an issue where the replication delay of a global read-only node could increase when the load on the primary nodes of a multi-master cluster was unbalanced.

• Fixed an issue where a newly created read-only node might fail to start on a multi-master cluster.

• Fixed an issue where executing a DDL on a multi-master cluster could cause the primary node to fail to restart.

Description

• Supports sampling statistics in the X-Engine to estimate the approximate compression ratio.

• Supports enabling the columnstore index feature on a global read-only node of a Multi-master Cluster (Database/Table).

• When using the columnstore index feature, allows users to ignore transaction isolation levels implicitly set by tools such as Metabase.

• Stats Manager supports automatically dispatching histogram collection tasks to read-only or hot replica nodes for automatic histogram updates.

• Supports using temporary tables when using the Hybrid Plan to accelerate wide table queries.

• Added the ability to enable semi-sync adaptive degradation and automatic recovery.

• Supports modifying the character set of a field using the INPLACE algorithm.

• SQL Trace supports opening configuration options for a limited number of read-only nodes. This means the sql_trace_type parameter can be configured as REPLICA_DEMAND, REPLICA_ALL, or REPLICA_SLOW_QUERY.

Description

• Improved the speed of handling connection requests by the Thread Pool in low-concurrency scenarios.

• Improved database performance in short-lived connection scenarios.

• Optimized the transaction mask calculation process in columnstore index (IMCI).

• When using the adaptive execution plan switching feature, if there is a high amount of cold data read IO, the maximum execution time for switching is automatically increased.

• Adjusted the maximum display length of query statements in SQL Trace.

• The columnstore index feature now includes a dynamic filter to improve the join efficiency between tables.

• When setting the value of the loose_optimizer_switch parameter using a SET_VAR hint in a HINT statement, a comma is no longer required at the end.

Description

• Fixed an incorrect sorting result issue when querying TopK data using the columnstore index (IMCI) feature. This occurred when sorting in descending order on a table with too many NULL values in the query statement.

• Fixed an issue where a read-only columnstore node could not recover columnstore data if it was started during the execution of a DDL involving a columnstore index table.

• Fixed a space leak issue caused by an exception during a shrink operation.

• Fixed an issue where concurrently modifying data structures during physical replication in the X-Engine caused a read-only node to become abnormal.

• Fixed an upgrade failure issue when upgrading an X-Engine from a read replica to a synchronous standby.

• Fixed an issue of excessive optimizer memory usage when using columnstore index in high-concurrency query scenarios.

• Fixed an issue where updating statistics on a partitioned table with a columnstore index could result in inaccurate statistics because only some partitions were sampled.

• Fixed an issue where a query statement containing a subquery that returned an empty result set could not be routed to a columnstore index node for execution.

• Fixed an issue where a CCL rule matched by a DIGEST value could become invalid after a cluster upgrade.

• Fixed an issue where querying the information_schema.tables table was abnormally slow when the database contained CSV or ORC external tables.

• Fixed a Flink CDC parsing failure issue caused by a duplicate table_id in the global Binlog of a Multi-master Cluster (Database/Table).

• Fixed an issue where resumable transaction data could not be queried when using the transaction breakpoint resumable upload feature because the parallel query feature was enabled.

• Fixed a failure to add a new RO node when executing the new database and table restoration process, caused by a too-low registration offset of the read-only (RO) node.

• Fixed an issue where a thread could not exit due to a full disk when executing the new database and table restoration process.

• Fixed an issue of high memory consumption when using Prepare Statement to handle many concurrent requests or execute complex queries.

• Fixed an issue where query results were incorrect when using columnstore index (IMCI) to query a view containing BIT type data.

• Fixed an issue where an insert operation on a KEY partitioned table would report an error "ERROR 1748 (HY000) Found a row not matching the given partition set". This was because the generated column (STORED GENERAGED COLUMN) was not considered when pruning the KEY partitioned table.

• Fixed an issue where new connections would still access a discarded old file due to frequent DDL and HA operations when archiving cold data as an IBD format file.

Category

Description

Release date

New features and performance optimizations

• Supports accessing the INFORMATION_SCHEMA view information of the column store from any node. By default, it aggregates and displays information from all column store nodes.

• Simplified the usage of parameters for automatic request distribution among row store and column store nodes.

• When a query statement does not support using the columnstore engine, you can set the use_imci_engine parameter to FORCED and then execute the query statement. The query result will then return the reason why the columnstore engine cannot be used.

• Supports using correlated subqueries in the Grouping clause and eliminating complex correlated subqueries.

• The global Binlog of a multi-master cluster supports GTID (global transaction identifier).

• Optimized the Buffer Pool expansion feature, such as automatically canceling preceding scale-in operations and timely expansion.

• Optimized the startup process of columnstore index (IMCI) to reduce the startup preparation time for column indexes.

• Supports adaptive routing capability for hybrid row-column queries.

• Improved the read performance of RO nodes in scenarios where a secondary index returns to the primary key.

2023-12-14

Bug fixes

• Fixed a node crash issue caused by an incorrect user-defined CONNECTION format when creating an OSS foreign table.

• Introduced a patch from official MySQL 8.0.17 to fix an autoinc rollback issue caused by modifying auto_increment_increment on the client.

• Introduced a patch from official MySQL 8.0.16 to fix a deadlock issue caused by an oversized Blob.

• Fixed an RO node crash issue caused by Runtime Apply.

• Fixed an issue where the RW node's dirty page flushing was blocked because it did not break its replication relationship with an RO node when the RO node's physical machine failed.

• Fixed an Autoinc rollback issue caused by table cache eviction due to an excessive number of tables after an HA.

• Fixed an issue where a read/write node in a multi-master cluster could not start if the cache center was restarted.

• Fixed an issue where a table could not be opened when switching the write point in a multi-master cluster.

• Fixed an issue of excessive memory usage during physical replication on an RO node.

• Fixed an issue where an outline on an AP node did not take effect if the length of the SQL token exceeded the value of the max_digest_length parameter.

• Fixed an issue where adding or deleting an outline could not be synchronized to an RO node after the RO node was upgraded.

• Fixed an issue of high response time for low-concurrency writes to tables in X-Engine.

• Fixed an issue where Hybrid Plan, when applied to an SQL statement with Order By and the output rows contained NULL, would report an error "table definition has changed".

Category

Description

Release date

Bug fixes

For clusters where no Statement Outline has been added, after a version upgrade, if a Statement Outline is added, outline matching is no longer affected by the character set.

2023-11-07

Category

Description

Release date

New features and performance optimizations

• Supports using the built-in stored procedure dbms_oss.delete_table_flie to delete unused files on OSS.

• Optimized the database and table restoration feature. For example, improved restoration speed and supported task failure rollback.

2023-10-31

Bug fixes

• Fixed an issue where a table with an IMCI column index could not be restored to a point in time.

• Fixed an issue where the GroupJoin operator consumed excessive memory in some scenarios, causing query failure.

• Fixed a RowID rollback failure issue when loading a checkpoint.

• Fixed an issue where a read-only (RO) node of the X-Engine could not report version reference information, causing space reclamation to fail.

• Fixed an issue where the optimizer could not accurately estimate the Join selectivity when joining an unsigned column with a signed column.

• Fixed an issue where some queries with OR predicates could not be executed because the query plan could not bind to the correct columns.

• Fixed an issue where a query terminated by the scheduler could cause a database restart.

• Fixed an IO performance degradation issue in columnstore index in late materialization scenarios.

• Fixed a rare issue where a multi-master cluster would crash because it did not release the metadata lock of a scheduled task in time.

• Fixed an issue where queries on tables in the X-Engine were slow because an index covering scan result could not be used after accessing the primary key index.

• Fixed an issue where a PolarDB cluster restart would interrupt the migration link when upgrading RDS for MySQL to PolarDB for MySQL with one click.

• Fixed an issue where deleting a table or index in X-Engine, due to a concurrency relationship with an occasional background SwitchMemtable operation, produced an extreme redo log sequence, causing the replica node to crash.

Category

Description

Release date

New features and performance optimizations

• Optimized the rate at which read-only (RO) nodes read Redo logs in large Blob scenarios.

• When calling built-in stored procedures, supports using uppercase stored procedure names.

2023-09-21

Bug fixes

• Fixed a node crash issue caused by adding backticks to the input parameters when calling some built-in stored procedures.

• Fixed a node crash issue that could occur when deleting the last CCL rule of the same SQL statement type, where only the keyword was filled in without a schema and table.

• Fixed a rare crash issue that could occur during the upgrade of a read-only node to a read/write node if there were file extension logs after the checkpoint on the read/write node when the new read-only node connected to it.

Category

Description

Release date

New features and performance optimizations

• Improved the speed of archiving cold data to OSS in CSV format.

• Optimized the cost model for equivalent non-covering index access.

• Added the parameter opt_non_covering_key_cost, to control whether to consider the initial positioning cost of a secondary index.

• Added the parameter worst_seek_records_threshold, to control whether to limit the maximum IO cost of equivalent non-covering index access during cost estimation.

• Added a timeout degradation mechanism to the Redo semi-sync feature.

2023-08-22

Bug fixes

• Fixed a crash issue with the JPPD optimization feature when a CTE contained a nested CTE and had multiple references.

• Fixed an issue where the Slave node did not trigger a reconnection mechanism when the primary node was shut down and restarted.

• Fixed a potential issue when using the JSON_REMOVE function in the syntax for updating BLOB type data.

• Fixed an issue where a node could restart abnormally due to accessing illegal memory during a transaction deadlock check.

Category

Description

Release date

Bug fixes

Fixed an issue of high CPU usage on low-specification clusters.

2023-07-13

Category

Description

Release date

New features and performance optimizations

• Added the SELECT INTO DEFAULT OSSOUTFILE statement.

• Added the parameter innodb_polar_import_tablespace_strict_mode with a default value of ON. This means that if the table schema contains a full-text index during a database and table restoration operation, an error will be reported.

• Supports querying data from a specified partition in a partitioned table using a columnstore index.

• SQL Trace supports sampling specific SQL statements.

2023-07-05

Bug fixes

• Fixed an issue where a regular user could delete sys.

• Fixed an abnormal node restart issue when performing database and table restoration on a partitioned table after an HA.

• Fixed an issue where the Standby node failed to switch to the master node during a point-in-time restore of a multi-master cluster.

• Fixed an issue where stored procedures, user-defined functions, scheduled tasks, views, and triggers were not isolated between different nodes in a multi-master cluster.

• Fixed an issue where stored procedures, user-defined functions, scheduled tasks, views, and triggers could not be switched between different nodes in a multi-master cluster.

• Fixed an issue where data in the Information_schema database could not be synchronized and updated if data was updated after the query cache feature was enabled for a multi-master cluster.

• Fixed a slow cleanup issue in a multi-master cluster with a very large number of data tables.

Category

Description

Release date

New features and performance optimizations

• Supports creating a columnstore index on a virtual column.

• Supports using the recycle bin feature for tables with a columnstore index.

• Supports creating a hybrid partition table.

• The federated query engine supports pushing down the Limit Offset syntax to the remote database for execution.

• The federated query engine supports pushing down all compatible conditions to the remote server for execution and returns only the columns required by the query.

• Added the primary key data slicing function call dbms_dts.show_parallel_ranges (<schema_name>, <table_name>, [<suggestion_ranges>]).

• Supports using the cold data archiving feature for tables with an auto-increment primary key.

• Optimized OSS query capability. After the elastic parallel query feature is enabled, if the system's loose_csv_max_oss_threads parameter value is greater than 1, a parallel query is performed directly.

• Optimized the error message content for incorrect OSS foreign table file formats.

• Supports adding the without validation option to the ALTER TABLE PARTITION BY RANGE syntax to quickly convert a standard table to a RANGE partitioned table.

• Supports converting a partitioned table with the 'UNIQUE CHECK IGNORE=1' attribute to a standard table using the ALTER TABLE REMOVE PARTITIONING syntax.

• Optimized an issue where the digest hash calculation was skipped for an executed statement when there was no Outline during the use of Statement Outline.

• Added the InnoDB on OSS feature.

• After updating Statement Outline and Concurrency Control on a multi-master cluster, other master nodes are automatically synchronized.

• Optimized the error message content for spelling errors.

• Optimized the write performance of the RW node after enabling StandBy semi-sync.

• Added the list default hash partitioned table type.

• Supports creating a LIST DEFAULT HASH type partitioned table using the CREATE LIKE syntax.

• The add partition syntax for creating a LIST DEFAULT HASH partitioned table supports using the without validation option to add a LIST partition .

• SQL Trace supports tracking the execution information of slow queries.

• Optimized the internal lock granularity of PolarFS, significantly improving performance in some high-load scenarios.

2023-06-12

Bug fixes

• Fixed an issue where OSS could not read JSON files.

• Fixed a Hint Plan generation failure issue in ps mode.

• Fixed an issue where the checkpoint LSN advancement did not meet expectations.

• Fixed an issue where view_definition was empty in the system view.

• Fixed a potential error when executing a query on an RO node while an ALTER TABLE ... IMPORT TABLESPACE operation was being performed on the RW node.

• Fixed an issue where data with \n could not be read after being archived by DLM.

• Fixed an issue where temporary files could remain on OSS after cold data archiving.

• Fixed an issue where only the first file might be read when reading CSV-formatted OSS cold data.

• Fixed an issue where an abnormal internal counter in the thread pool could cause SQL request stacking during concurrent recording of slow query logs.

• Fixed a potential auto-increment column conflict issue during concurrent INSERT command execution in a multi-master cluster.

• Fixed a replication delay issue when replaying Redo logs of a wide table on an RO node.

• Fixed an incorrect query result issue when using the Limit Offset pushdown optimization feature with a query statement containing a window function.

• Fixed an issue where a lock needed to be held for the entire duration when executing a CHECK TABLE statement in a multi-master cluster.

• Fixed an issue where the source file on OSS could not be deleted when executing the ALTER TABLE engine = csv storage oss, force; command.

• Fixed a cluster crash issue when dynamically adjusting the cache of a multi-master cluster.

• Fixed an issue where a multi-master cluster could not export a tablespace.

• Fixed an incorrect pushdown issue for tables with NULL fields when the engine_condition_pushdown parameter was set to true.

• Fixed a potential incorrect result set issue when using the Fast Query Cache feature with Time and timestamp comparisons.

• Fixed a potential performance regression issue in high-concurrency update scenarios after dynamically enabling PolarTrans.

Category

Description

Release date

New features and performance optimizations

• Columnstore index supports instant column addition by rebuilding columnstore index data.

• Columnstore index supports the REGEXPR function.

• Columnstore index supports pruning optimization based on IS NULL and IS NOT NULL conditions.

• Multi-master Cluster (Database/Table) supports using the cold data archiving feature.

• When performing a minor engine version update on a Multi-master Cluster (Database/Table) edition cluster, supports parallel upgrade of each node in the cluster.

• Multi-master Cluster (Database/Table) supports persisting statistics.

• Optimized the Optimizer trace when OSS condition pushdown is enabled for an OSS foreign table.

• Binary logs are not recorded during manual and automatic archiving of cold data.

• Global consistency (high-performance mode) adds the Innodb_scc_timeout_count parameter to record the number of timeouts during a query.

• Global consistency (high-performance mode) adds the Innodb_scc_timeout_degrade_count parameter to record the number of times a query times out and switches to an asynchronous query.

• Supports querying gtid information on a read-only node.

• Hint Plan adds the hint types BNL, BKA, and MRR.

• Fast Query Cache adds the variable qc_weak_read_time to read non-real-time cache.

• The Statement Outline feature adds the interface dbms_outln.add_optimizer_outline('schema', 'hints', 'query') to add Optimizer Hints.

• Added the Auto Plan Cache feature.

• SQL Sharing now displays plan_extra, ref_by, and extra in JSON format.

• Displays the execution plan in the extended data of the slow log.

• When using Statement Outline, if the specified Index Hints do not exist, it reports a warning instead of an error.

• SQL Sharing adds the variable by_expr_ignore_charset. After setting by_expr_ignore_charset to ON, it eliminates digest differences generated by statements with Order By or Group By under different character sets. However, after setting it to ON, all digest values generated in this mode will be different from the digest values generated before it was set to ON.

• Optimized memory consumption when scanning tables in the recycle bin.

• Provides multi-version engine capability.

2023-04-14

Bug fixes

• Fixed an issue where the database node would crash when performing a check operation on an empty OSS foreign table.

• Fixed an occasional error when executing a query on a global read-only node of a Multi-master Cluster (Database/Table).

• Fixed an issue where the metadata tablespace size was not synchronized in a Multi-master Cluster (Database/Table).

• Fixed an issue where custom table data in the system tablespace would be rolled back in a Multi-master Cluster (Database/Table).

• Fixed an issue where undo logs could not be recycled after enabling the undo log recycling feature.

• Fixed a database connection failure issue due to a read timeout when establishing physical replication on a read-only node.

• Fixed a node crash issue when executing an ADD INDEX or DROP INDEX statement after executing ALTER TABLE DISCARD tablespace in a Multi-master Cluster (Database/Table).

Category

Description

Release date

New features and performance optimizations

• The columnstore index feature supports the MySQL DECIMAL data type.

• When creating a columnstore index, supports sorting data by a specified column to improve pruning efficiency.

• The columnstore index feature supports data scanning based on files in ORC or PARQUET format stored in OSS.

• Cost-based computation supports pushing down the outer join conditions of a materialized derived table into the derived table.

• Optimized the optimizer's ability to estimate the number of scanned rows for join columns in an equivalence derivation scenario.

• The optimizer's join elimination feature supports eliminating the inner table of a left join (including inner tables with multi-level nested joins and Semi-Joins).

• Supports the CAST(... AS INT) syntax, with behavior identical to CAST(... AS SIGNED INT) .

• Multi-master clusters support multi-master backup.

2023-02-21

Bug fixes

• Fixed an issue where the database performance slowed down in long-running transaction scenarios.

• Fixed a compatibility issue between global consistency (high-performance mode) and PQ.

• Fixed an issue where using the transparent tablespace transport feature in a multi-master cluster would block other nodes from executing DDL.

• Fixed an issue where the free size of a table would, with a certain probability, be displayed as 4 PB.

• Fixed an issue where some nodes could not start when multiple primary nodes of a multi-master cluster were restarted in parallel.

• Fixed a cluster crash issue when switching the write point for full-text indexes and columnstore indexes in a multi-master cluster.

• Fixed a database cache access anomaly when performing a copy DDL on a table with a Foreign Key.

• Adjusted physical replication connection-related timeout parameters to avoid stuttering issues during physical machine failures.

• Fixed an issue where table locks could not be recovered during a multi-master cluster crash recovery.

• Fixed a potential node crash issue when accessing the information_schema.tables table when adding a new node.

• Fixed a dirty page loss issue during the multi-master cluster crash recovery phase.

• Fixed a cluster crash issue when switching the write point after deleting a database in a multi-master cluster.

• Fixed an issue where using the TempTable engine could lead to illegal memory access when the thread pool was enabled.