PolarDB:PolarDB for MySQL 5.7

Category

Description

Release date

New feature and performance optimization

• The data of a PolarDB for MySQL 5.6 cluster can be physically migrated to a PolarDB for MySQL 5.7 cluster.

• New tables can be automatically encrypted by using transparent data encryption (TDE) after TDE is enabled.

• Fine-grained data modification tracking is supported to improve the performance of global consistency (high-performance mode).

• Custom temporary tables can be created in memory by using the MEMORY or MyISAM engine on read-only nodes.

• The original read-only max_digest_length parameter can be manually configured.

• The simulated asynchronous I/O (AIO) performance is improved.

• An outline can be applied to optimize the execution of a query even if the database name is not specified in the outline.

• The number of input/output (I/O) threads in the cluster can be automatically adjusted without a restart of the cluster.

2024-08-30

Fixed issue

• The following issue is fixed: A lock conflict occurs during a dynamic scale-in of the buffer pool.

• The issues related to the purge prefetching of the blob-type data are fixed.

• The following issue is fixed: Autoinc values in a partitioned table are rolled back after an HA switchover.

• Performance issues that occur during the serialization of complex objects into the JSON format are fixed.

• The following issue is fixed: The precision of floating-point numbers in JSON is insufficient.

• Deadlocks due to frequent record locks caused by unnecessary unique key constraints are fixed.

• The following issue is fixed: Access of user threads to free pages is delayed when the buffer pool of a read-only node is full.

• The following issue is fixed: The Autoinc values roll back to 0 after all data of the table is cleared.

Category

Description

Release date

New feature and performance optimization

• The preemptive DDL feature is supported to resolve MDL acquisition failure during DDL synchronization.

• Thread priorities in the PolarFileSystem (PFS) are optimized to ensure that transactional requests are processed at the earliest opportunity. This improves the performance of DDL statements in PolarDB for MySQL.

• The page read performance of read-only nodes on the PFS is enhanced.

2024-07-01

Fixed issue

• The following issue is fixed: An exception may occur in an MDL when the number of connections exceeds 65,536.

Category

Description

Release date

New feature and feature optimization

• The message that indicates the failure of read-only nodes to acquire the metadata lock (MDL) is optimized.

• The polar_only_allow_discard_with_admin parameter is added to prohibit users from performing DISCARD operations. The default value is ON. Users can configure this parameter based on their business requirements.

• The fast query cache feature is optimized to include information about the number of returned rows. For queries that hit the cache, the number of returned rows is displayed in the audit log.

2024-05-27

Fixed issue

• The following issue is fixed: System operations and maintenance (O&M) activities affect the slow_queries status metric.

• The following issue is fixed: A comma (,) must be added at the end when SET_VAR hint is used to configure the loose_optimizer_switch parameter.

Category

Description

Release date

New feature and feature optimization

• After the global transaction identifier (GTID) feature is enabled, temporary tables can be created and deleted in transactions.

• The Variable-Setting hint syntax is supported in the global consistency (high-performance mode) feature.

• The performance of the thread pool is optimized when multiple threads of binary logs are concurrently written.

2024-03-27

Fixed issue

• The following issue is fixed: When a child process is forked, TDE is incompatible with the RDMA memory.

• The following issue is fixed: Subsequent scale-out requests cannot be processed after the buffer pool is scaled in because large SQL statements occupy a large number of pages for a long time.

• The following issue is fixed: Users must restart the node in case of autoinc rollback.

• The following issue is fixed: A deadlock occurs in parallel DDL scenarios when DDL read-ahead is enabled.

• The following issue is fixed: A new INSERT request is autoinc duplicate key after a DDL statement that modifies a partition is executed on a partition table.

Category

Description

Release date

New feature and performance optimization

The speed at which the thread pool feature processes connection requests in low-concurrency scenarios is improved.

2024-01-31

Fixed issue

• The following issue is fixed: Nodes may crash when the logical prefetching feature is used.

• The following issue is fixed: Memory leak issues occur when the logical prefetching feature is used.

• The following issue is fixed: When a new process is initiated for restoring individual databases and tables, the thread cannot exit due to the full disk issue.

• The following issue is fixed: When a new process is initiated for restoring individual databases and tables, the read-only node fails to be added because the redo log on the primary node is cleared.

Category

Description

Release date

Fixed issue

• The following issue is fixed: When autoinc_persist is enabled, autoinc rollback occurs after an HA switchover action.

• The following issue is fixed: Autoinc rollback occurs on the primary node or read-only nodes after repeated HA switchover actions.

• The following issue is fixed: After a degrade operation is performed on an primary node, the physical replications on the standby nodes cannot be reconnected.

2023-12-11

Category

Description

Release date

New feature and performance optimization

• The database and table restoration feature is optimized. For example, databases and tables can be restored more quickly, and failed tasks can be rolled back.

• The log output feature is optimized: Unnecessary error logs for the fil_space_extend function are deleted.

2023-11-23

Fixed issue

• The following issue is fixed: When the physical machine of a read-only node fails, the primary node does not terminate the data replication under special circumstances. As a result, the primary node cannot flush data.

• The following issue is fixed: Autoinc rollback caused by table cache eviction occurs due to a large number of tables after an HA switchover action.

• The following issue is fixed: When an ApsaraDB RDS for MySQL instance is upgraded to a PolarDB for MySQL cluster, the migration link is interrupted due to the restart of the cluster.

• The following issue is fixed: Abnormal checkpoints occur after an UNDO TRUNCATE statement is executed.

• The following issue is fixed: When a read-only node is started, it is connected to the primary node not by using the polarfs dir fast scan API.

Category

Description

Release date

Fixed issue

• The following issue is fixed: Deadlocks occur when the cross-page logical prefetching conflicts with the merge sub-operation of a split-merge operation (SMO).

• The following issue is fixed: The data of an InnoDB temporary table cannot be flushed.

• The following issue is fixed: Nodes may crash if only keywords are specified but not the table schema and table name in the last rule when multiple concurrency control (CCL) rules are configured to delete the same type of SQL statements.

2023-09-14

Category

Description

Release date

Fixed issue

The following issue is fixed: When the page cleaner thread has a large amount of dirty data to flush but only a small percentage of the available memory can be utilized for dirty pages, the write performance of the database is compromised.

2023-08-28

Category

Description

Release date

Fixed issue

The following issue is fixed: When audit log is enabled, SQL latency increases due to loaded local disks.

2023-08-02

Category

Description

Release date

New feature and performance optimization

• The innodb_polar_import_tablespace_strict_mode parameter is added. The default value is ON, which specifies that an error is returned if a database or table to be restored contains full-text indexes.

• The innodb_polar_scc session-level parameter is added to specify whether to enable strong consistency for current connection requests. When the global consistency (high-performance mode) feature is enabled and the innodb_polar_scc parameter is set to true, global consistency takes effect.

2023-07-23

Fixed issue

• The following issue is fixed: A standard account can delete the performance_schema database.

• The following issue is fixed: Checkpoint LSNs are not pushed down as expected.

• The following issue is fixed: An error may be returned when the ALTER TABLE ... IMPORT TABLESPACE statement is executed on a read-write node and a query is performed on a read-only node at the same time.

• The following issue is fixed: When the primary node of an instance is stopped and then restarted, the secondary nodes of the instance cannot be reconnected to the primary node.

• The following issue is fixed: When a large value is specified for the group_concat_max_len parameter, an arithmetic overflow occurs, which causes an incorrect result of the GROUP_CONCAT function.

• The following issue is fixed: Contention for mutex locks occurs in concurrent update scenarios after PolarTrans is enabled.

• The following issue is fixed: After the undo log reclamation mechanism is triggered on the primary node, read-only nodes crash if the nodes are connected to the database to read data.

Category

Description

Release date

Fixed issue

The following issue is fixed: When slow query logs are concurrently recorded, a large number of SQL requests are piled up due to the abnormality of internal counter in the thread pool.

2023-06-14

Category

Description

Release date

New feature and performance optimization

• DDL and LOCK TABLE statements can be recorded to the system tables.

• The thread pool feature is compatible with CCL. After the thread pool feature is enabled and if a connection is blocked due to CCL, the connection is marked within the thread pool. The thread pool feature creates additional threads to execute the SQL statements from other connections.

• If the index hints specified for statement outline do not exist, only an alert is reported instead of an error.

• The show full tables from db_name like table_name syntax is optimized.

• The Innodb_scc_timeout_count parameter is added for global consistency (high-performance mode) to record the number of timeout events during queries.

• The Innodb_scc_timeout_degrade_count parameter is added for the global consistency (high-performance mode) feature to record the number of times that a query times out and is converted to an asynchronous query.

• SQL statements are optimized to perform digest hash calculation when no outlines are available after Statement Outline is enabled.

• The efficiency of truncating temporary tables is optimized.

• Global transaction identifiers (GTIDs) can be queried only on read-only nodes.

• The lock granularity of PolarFS is optimized to significantly improve performance in some high-load scenarios.

2023-05-08

Fixed issue

• The following issue is fixed: Created events cannot be deleted.

• The following issue is fixed: A read timeout error occurs and the database cannot be connected when physical replication is established on a read-only node.

• The following issue is fixed: The undo logs cannot be truncated.

• The following issue is fixed: When the redo logs of a wide table is used on read-only nodes, replication latency occurs.

Category

Description

Release date

Fixed issue

• The following issue is fixed: Hot standby nodes may unexpectedly crash when the global consistency (high-performance mode) feature is enabled.

• The following issue is fixed: PolarTrans logs cannot be reclaimed after the primary node is restarted.

2023-03-25

Category

Description

Release date

Fixed issue

• The following issue is fixed: The AUTO_INCREMENT value in the CREATE TABLE statement is lost after a switchover between the primary and read-only nodes.

• The following issue is fixed: After a cluster is started, the configurations of the innodb_buffer_pool_in_core_file parameter in the my.cnf file of the cluster do not take effect.

• The following issue is fixed: Communication with the TDE extension is slow when large memory specifications are used.

2023-02-27

Category

Description

Release date

New feature and performance optimization

• The semicolon (;) at the end of an SQL statement of the query parameter in the following built-in stored procedures is ignored: dbms_ccl.add_ccl_digest_rule, dbms_outln.add_index_outline, and dbms_outln.add_optimizer_outline.

• All types of UNION clauses are supported for Statement Outline.

• The timeout parameter related to the physical replication connection is modified to prevent issues when the physical machine where the primary node is deployed fails.

• Thread pool optimizations:

  • The thread_pool_high_priority_users parameter can be used to configure a list of high-priority users. The thread pool preferentially allocates threads to requests from high-priority users.

  • The bypass mechanism is added. The bypass_thread_pool_ips parameter can be specified to grant special privileges to specific IP addresses. When the IP addresses specified in the bypass_thread_pool_ips parameter is used to connect to the database, the thread pool is bypassed.

  • The timeout mechanism for slow queries is added. If the thread pool counts the overall number of threads, the number of threads in the slow query state is ignored, so that the overall number of threads is not limited by the number of slow query threads.

2023-01-06

Fixed issue

The following issue is fixed: The database performance may be degraded in long-running transaction scenarios.

Category

Description

Release date

New feature and performance optimization

• The Innodb_scc_timeout_count parameter is added for the global consistency (high-performance mode) feature to record the number of timeouts that occur during a query.

• The Innodb_scc_timeout_degrade_count parameter is added for the global consistency (high-performance mode) feature to record the number of times that a query times out and is converted to an asynchronous query.

2022-11-11

Fixed issue

• The following issue is fixed: The physical replication module on the primary node occupies additional memory.

• The following issue is fixed: Undo logs cannot be cleared in special scenarios.

• The following issue is fixed to improve system stability in high-concurrency scenarios: Keepalive check control fails in special scenarios.

Category

Description

Release date

Fixed issue

The following issue is fixed: Read-only nodes crash when the hash values of file names conflict.

2022-09-17

Category

Description

Release date

New feature and performance optimization

• The global consistency (high-performance mode) feature is added. After the global consistency (high-performance mode) feature is enabled for a PolarDB cluster, read-only nodes in the cluster provide strong consistency for read operations. For more information, see Overview.

• The read-ahead mechanism for DDL statements is optimized to improve the efficiency of creating indexes.

• Optimized DDL sorting: Two-way merging is changed to multiple-way merging to reduce I/O operations and improve the performance of creating indexes by executing DDL statements.

• The performance of the thread pool in high-concurrency environments is enhanced.

2022-08-30

Fixed issue

The following issue is fixed: Slow logs are not recorded when a database is deleted.

Category

Description

Release date

New feature and performance optimization

The read-ahead mechanism for DDL statements is supported, which improves the efficiency of index creation.

2022-06-17

Fixed issue

• When the optimizer selects indexes that can accelerate the execution of ORDER BY clauses, it takes the costs for obtaining data from tables into consideration.

• The following issue is fixed: When a large number of dirty pages exist in the buffer pool and page buffer pool flushing is slow, the least used page cannot be released.

Category

Description

Release date

New feature and performance optimization

The latency of physical replication between primary and secondary nodes is reduced to milliseconds.

2022-05-20

Fixed issue

• The issue that causes high latency when physical replication is performed during a time period of low workloads in the cluster is fixed.

• The process for selecting indexes is improved. Indexes that can provide ordering information to accelerate the execution of ORDER BY clauses are prioritized.

• The following issue is fixed: The optimizer fails to choose a better path to the range of an index for some GROUP BY statements.

• The following issue is fixed: When the primary node establishes a replication relationship with read-only nodes, the primary node fails to immediately apply the checkpoint policy.

Category

Description

Release date

New feature and performance optimization

During SQL parsing, OF, COMMIT_ON_SUCCESS, ROLLBACK_ON_FAIL, and TARGET_AFFECT_ROW are changed from reserved keywords to non-reserved keywords.

2022-04-07

Category

Description

Release date

New feature and performance optimization

• The hot row optimization feature is supported. For more information, see Hot row optimization.

• The hot standby node feature is supported to further optimize the high availability efficiency.

2022-01-14

Category

Description

Release date

Fixed issue

The policy of manually triggering checkpoints is optimized.

2021-12-14

Category

Description

Release date

New feature and performance optimization

The fast query cache feature is added. For more information, see Fast query cache.

2021-11-17

Category

Description

Release date

Fixed issue

• Optimizer hints are provided for users to determine whether to force merge views. For more information, see MySQL official document.

• The SELECT FOR UPDATE WAIT N syntax is supported. N specifies the number of seconds during which a row remains locked. The timeout applies only to locking a single row. To lock multiple rows for a query, the system detects the timeout only based on a single row instead of multiple rows. If the lockout time exceeds the specified N value, the system returns the following error message: Lock wait timeout exceeded; try restarting transaction.

2021-10-19

Category

Description

Release date

New feature and performance optimization

A read-only node of a secondary cluster in a global database network (GDN) can be promoted to be the primary node by running the alter polar to slave command. This implements the high availability of clusters.

2021-09-10

Category

Description

Release date

New feature and performance optimization

• Multiple connections can be established between the primary cluster and synchronization nodes, which improves the throughput of physical replication.

• Binary logs can be obtained from read-only nodes. For more information, see Remotely obtain and parse binary log records from a cluster of PolarDB for MySQL.

2021-08-27

Fixed issue

• The statement concurrency control rules where databases and tables are specified cannot match actual conditions.

• The process of loading redo logs is accelerated for the read-only nodes and the secondary cluster, which improves the synchronization efficiency of the primary node.

Category

Description

Release date

New feature and performance optimization

The polar_replica_work_on_nonblock_mdl_mode parameter is added. If this parameter is set to on, the uncommitted transactions that use the READ UNCOMMITTED or READ COMMITTED isolation level on read-only nodes no longer block DDL operations on the primary node. In addition, atomicity, consistency, isolation, and durability (ACID) is no longer ensured for the operations of creating tables on the read-only nodes.

2021-07-08

Category

Description

Release date

New feature and performance optimization

This version is compatible with MySQL 8.0 and supports the SELECT FOR UPDATE, SHARE SKIP LOCKED, and NOWAIT statements. For more information, see MySQL 8.0.

2021-06-15

Fixed issue

Data in temporary tables in InnoDB databases can be saved to disks by using read-only nodes.

Category

Description

Release date

Fixed issue

• When data is retrieved from multiple tables, the storage engine can be quickly launched.

• The following issue is fixed: The database crashes when an update operation on a virtual column is rolled back.

2021-05-13

Category

Description

Release date

New feature and performance optimization

A feature is added to decide whether the primary node must apply the checkpoint policy immediately when it establishes a replication relationship with read-only nodes.

2021-04-19

Fixed issue

• KICKOUT is used as a non-reserved keyword.

• The following issue is fixed: The encryption information in tables that are encrypted by TDE may be lost during a failover. This results in decryption errors.

Category

Description

Release date

New feature and performance optimization

• The parallel DDL feature is supported. This feature accelerates the execution of DDL statements. For more information, see Parallel DDL.

• The innodb_buffer_pool_in_core_file parameter is added. This way, buffer pools can be removed from core files.

• The following index hints are supported. This allows the optimizer to use or ignore the specified indexes when the optimizer processes queries.

  • GROUP_INDEX or NO_GROUP_INDEX: The system uses or ignores the specified indexes for an index scan that contains GROUP BY operations.

  • INDEX or NO_INDEX: The system forces the server to use or ignore the specified indexes.

  • JOIN_INDEX or NO_JOIN_INDEX: The system forces MySQL to use or ignore the specified indexes regardless of what method is used to query data.

• The following join order hints are supported. This allows the optimizer to join tables in a valid order.

  • JOIN_FIXED_ORDER: The system forces the optimizer to join tables by using the order in which the tables appear in the FROM clause.

  • JOIN_ORDER: The optimizer is configured to join tables by using the specified order. This hint applies to named tables. The optimizer may place tables that are not named anywhere in the join order, including between specified tables.

  • JOIN_PREFIX: The optimizer is configured to join the first several tables in the join plan by using the specified order. This hint applies to named tables. The optimizer places all other tables after named tables.

  • JOIN_SUFFIX: The optimizer is configured to join the last several tables in the join plan by using the specified order. This hint applies to named tables. The optimizer places all other tables before named tables.

2021-03-10

Fixed issue

• The mapping of token numbers is rearranged and some tokens are reserved. This fixes the issue that the digest hash value of a statement is changed after a new token is inserted.

• The following issue is fixed: A configuration update may fail because read-only nodes cannot be registered on the primary node.

• The following issue is fixed: A cluster abnormally exits due to the creation or deletion of temporary tables during a stored procedure after the session_track_temporary_tables system variable is enabled.

Category

Description

Release date

Fixed issue

The following issue is fixed: The SHOW PROCESSLIST statement in some scenarios returns an invalid result.

2021-02-22

Category

Description

Release date

Fixed issue

The following issue is fixed: The 1032 error code and the Can't find record in TABLE error message are reported when a query optimized by Index Merge is performed on a read-only node.

2021-02-10

Category

Description

Release date

Fixed issue

• The following issue is fixed: Dirty page flushing cannot be performed as expected on read/write nodes.

• The following issue is fixed: The primary zone cannot be changed for a cluster that performed a failover.

• The following issue is fixed: The database crashes if Thread Descriptor (THD) is invoked when the SHOW PROCESSLIST statement is being executed.

2021-02-02

Category

Description

Release date

New feature and performance optimization

• The Statement Queue feature is added. For more information, see Statement queue.

• The Instant ADD COLUMN feature is supported. This feature can instantly add columns to a table. For more information, see Instant ADD COLUMN.

• The Returning feature is added. For more information, see Returning.

• The default value of the innodb_adaptive_hash_index parameter is changed from ON to OFF.

2021-01-29

Fixed issue

The following issue is fixed: The system tables of a PolarDB cluster are lost when the engine version of the cluster is upgraded.

Category

Description

Release date

Fixed issue

• The following issue is fixed: Read-only nodes crash when data is queried from the read-only nodes after the key of the primary node is modified.

• The following issue is fixed: Read-only nodes crash when partitioned tables are analyzed on the read-only nodes.

• The following issue is fixed: The new primary node crashes due to tablespace inconsistency after a failover is performed.

2020-12-31

Category

Description

Release date

Fixed issue

• The following issue is fixed: Data cannot be inserted into temporary tables after a failover is performed.

• The following issue is fixed: A read-only node becomes unavailable when a temporary tablespace is extended on the read-only node.

• The following issue is fixed: The simulated AIO does not work as expected.

2020-11-17

Category

Description

Release date

New feature and performance optimization

Transparent data encryption (TDE) is supported.

2020-10-21

Fixed issue

• The following issue is fixed: Statistics are inconsistent between the primary node and read-only nodes.

• The following issue is fixed: An error message is returned by the SHOW PROCESSLIST statement.

Category

Description

Release date

Fixed issue

• The following issue is fixed: IDB files cannot be created after a migration from ApsaraDB RDS for MySQL 5.7.

• CF_STATUS_COMMAND commands cannot be run during a failover process. This is to ensure that the failover process succeeds.

• The following issue is fixed: The primary node becomes unavailable due to page contention that is caused by updating the aggregation thread and the truncate logic in the background.

2020-08-28

Category

Description

Release date

Fixed issue

• The following issue is fixed: MaxScale becomes inaccessible due to the missing code of some exclusive commands in PolarDB.

• The following issue is fixed: The cluster that is restored from the cluster recycle bin is unavailable because no tablespaces are found.

• The following issue is fixed: The unmount PFS error occurs if one or more files are still open when the promote replica command is run.

2020-07-30