This topic describes the updates and features of the PolarDB for MySQL database kernel.

The following figure shows the position of the PolarDB database engine in the PolarDB architecture.

1

For more information about the architecture and features of PolarDB, see Architecture.

Query the kernel version number

You can use the following methods to query the kernel version number of a cluster:

  • Log on to the PolarDB console. On the Overview page of the cluster whose kernel version you want to query, check the value of the Kernel Version parameter.1
  • If the value of the Kernel Version parameter is not displayed in the console, execute the show variables like "%polardb_version%"; or show variables like '%rds_release_date%'; statement to query the kernel version number.
    Note
    • You can execute only the show variables like '%rds_release_date%'; statement to query the kernel version number of a PolarDB for MySQL 5.6 cluster.
    • For more information about the version number of the PolarDB database kernel, see Version of the PolarDB database kernel.

PolarDB for MySQL 8.0 release notes

The following table describes the versions and features that are supported by PolarDB for MySQL 8.0.

Minor version Revision version Release date MySQL minor version with full compatibility Released feature
8.0.2.x

In public preview

8.0.2.1.4 2020-12-01 8.0.18 The Index Hints feature of MySQL 8.0.20 is supported. For more information, see Index Hints.
8.0.2.1.3 2020-10-26
  • The cost estimation model for hash joins is improved.
  • Parallel hash joins are supported. The build phase and the probe phase can be executed in parallel.
8.0.2.1.2 2020-09-27 The latest volcano iterator executor is supported.
8.0.2.1.1 2020-08-26 Fixes several known issues.
8.0.2.1.0 2020-07-22
8.0.1.x 8.0.1.1.14 2021-04-23 8.0.13
  • The index maintenance process of PolarDB for MySQL has been optimized for a shorter response time on read-only nodes in scenarios where a large number of tables exist in the database, such as providing SaaS.
  • The synchronized interconnection among global database networks (GDNs) has been optimized to reduce the consumption of CPU resources by synchronized threads and improve the speed of CPUs with eight or less cores when they synchronize redo logs from clusters.
  • The AutoDop policy of the parameters for controlling the degree of parallelism has been optimized to avoid affecting the use of PARALLEL HINT and force_parallel_mode.
8.0.1.1.13.2 2021-04-19 Fixes a known issue.
8.0.1.1.13.1 2021-04-08 Fixes a known issue.
8.0.1.1.13 2021-03-30
  • Read-only nodes support the polar_use_statement_mdl_on_replica parameter. This setting prevents transactions of the read committed (RC) isolation level on read-only nodes from blocking DDL operations on the primary node. When read transactions on read-only nodes are concurrent with DDL operations on the primary node, read transactions will read different table definitions. For example, if an ADD COLUMN operation on the primary node is performed between two statements of a read transaction on the read-only node, the second statement on the read-only node reads more columns than the first statement.
  • Unnecessary equality conditions on indexed values in a query are removed. This pushes down the Limit offset statement when the ORDER BY and LIMIT clauses are used.
  • The dbms_stats.gather_stats(timeout, workers) command is added. You can update the outdated histogram by manually executing this command or using scheduled events.
  • You can query data of the log_version, log_id, origin_start_time, and rds_ext fields in the mysql.slow_log table.
8.0.1.1.12.2 2021-03-12 Fixes several known issues.
8.0.1.1.12.1 2021-03-02 The write performance of X-Engine in PolarDB Archive Database is optimized when you continuously import data to multiple tables.
8.0.1.1.12 2021-02-20
  • The auto_dop_low_degree_cost parameter is supported in the policies for the degree of parallelism. This parameter is used to determine how to automatically specify the degree of parallelism for parallel queries. For more information, see DOP policy.
  • The restore_table command is supported to allow you to quickly restore tables from the recycle bin. For more information, see Recycle bin.
  • You are allowed to retrieve binary logs from read-only nodes. For more information, see Remotely obtain and parse binary logs from a PolarDB for MySQL cluster.
  • The key information such as in_memory can be printed in opt trace. This allows you to quickly locate causes when an exception occurs to an execution plan.
8.0.1.1.11 2021-01-29
  • allows you to specify the parallel_degree_policy parameter to set the degree of parallelism (DOP) policy for parallel queries.
  • The SET GLOBAL statement is supported to specify the max_digest_length parameter. This parameter is used to limit the length of recognizable statements.
8.0.1.1.10 2021-01-12
  • The implicit sorting feature of Group By is supported.
  • The hot row optimization feature is supported. For more information, see Hot row optimization.
  • The physical replication optimization feature of Data definition language (DDL) operations is supported. For more information, see DDL physical replication optimization.
  • The async metadata lock replication feature is supported. For more information, see Async Metadata Lock Replication.
  • The cluster performance is optimized in scenarios where a large number of tables exist in a cluster.
8.0.1.1.9 2020-12-18 The SPM and PLAN keywords are removed.
8.0.1.1.8 2020-12-09
  • The multi-plan mode is supported in the execution plan manager.
  • The rds_ap_threshold system variable is added to block the requests that cause the system to scan a large amount of data. These requests are evaluated by the optimizer.
  • The efficiency of writing dirty pages to disks on the primary node is improved.
  • A multi-shard writing mechanism for redo logs is supported.
8.0.1.1.7 2020-11-16
  • The efficiency of scanning driven tables in parallel is improved in scenarios such as Join queries.
  • Remaining binary logs can be cleared when the binary logging feature is disabled.
  • Interrupted physical replications of secondary nodes can be automatically detected and then resumed.
  • The efficiency of the failover between the primary node and read-only nodes is improved.
  • Quick startup is supported for clusters that contain a large number of tables. This way, data files can be rapidly scanned.
8.0.1.1.6 2020-09-21
  • The compatibility between the SPM and parallel queries is improved.
  • The efficiency of merging and sorting parallel queries is improved.
  • DELETE operations can be pushed down to InnoDB.
  • The Commit Timestamp (CTS) feature of PolarDB is supported.
8.0.1.1.5 2020-08-19
  • A feature is added to check whether the primary node immediately applies the checkpoint policy when it establishes a replication relationship with read-only nodes.
  • Simple range queries are supported. Range queries can be processed in InnoDB.
8.0.1.1.4 2020-07-04
  • DDL statements can be executed in parallel. This improves the efficiency of DDL statements.
  • The length of each queue for the simulated AIO can be dynamically adjusted.
  • Data in an FTS index cache can be synchronized between the primary node and read-only nodes. This ensures cache coherence.
8.0.1.1.3 2020-05-29 The parallel query feature is improved.
8.0.1.1.2 2020-04-09
  • The recycle bin feature is supported. This prevents data loss that is caused by unintended DROP operations. For more information, see Recycle bin.
  • The operations that are performed on temporary tables compromise the performance. This version is optimized to reduce the negative impact of these operations on performance.
  • The parallel query feature is improved.
8.0.1.1.1 2020-03-28
  • The thread pool feature is supported. For more information, see Thread pool.
  • Global database networks (GDNs) are supported. For more information, see Global database networks.
  • The parallel query feature is improved.
8.0.1.1.0 2020-02-05
  • The cost estimation model for parallel queries is improved.
  • The thread scheduling capabilities of parallel query executors are enhanced.
8.0.1.0.6 2020-01-01 Fixes several known issues.
8.0.1.0.5 2019-12-03 Diagnosis information about the execution plans of parallel queries is added. This ensures that the execution plans are easy to read.
  • 8.0.2.1.4(20201201)
    Category Description
    New features and optimized performance The Index Hints feature of MySQL 8.0.20 is supported.
    Fixed issues
    • The stability issue of parallel hash joins is fixed.
    • The following issue is fixed: The estimated number of rows for each index key in statistics is miscalculated.
    • The following issue is fixed: An error occurs in processing empty data when shared temporary tables are scanned in parallel.
    • The following issue is fixed: An error occurs when you use the official REGEXP_REPLACE function for data manipulation.
    • The following issue is fixed: A crash occurs when a complex query contains subquery constants.
    • The following issue is fixed: The filesort returns incorrect data.
  • 8.0.2.1.3(20201026)
    Category Description
    New features and optimized performance
    • The cost estimation model for hash joins is improved.
    • Parallel hash joins are supported. The build phase and the probe phase can be executed in parallel.
    Fixed issues
    • The following issue is fixed: The wrong number of parallel partitions in a parallel execution plan is displayed.
    • The following issue is fixed: A crash occurs for parallel subqueries.
    • The following issue is fixed: Multiple worker threads cannot return random results when the RAND() function is used in parallel queries.
  • 8.0.2.1.2(20200927)
    Category Description
    New features and optimized performance PolarDB supports the latest volcano iterator executor for parallel queries.
  • 8.0.2.1.1(20200826)
    Category Description
    New features and optimized performance The write performance of clusters is optimized.
    Fixed issues The memory leak issue is fixed.
  • 8.0.2.1.0(20200722)
    Category Description
    New features and optimized performance
    • The EXPLAIN statement can be executed to query the execution plan of parallel queries. After you specify FORMAT=TREE in the EXPLAIN statement, the execution plan is displayed in the TREE format.
    • Hash joins are supported for parallel queries. The performance of JOIN operations that do not use indexes is improved. For more information, see Hash joins in parallel queries.
    • The Resource Manager feature is added to monitor CPU and memory resources. For more information, see Resource Manager.
    • The Performance Agent feature is added to collect statistics about the performance metrics of PolarDB for MySQL clusters. To collect the performance metrics, you can query the PERF_STATISTICS in-memory table. For more information, see Performance Agent.
    • The warm buffer pool feature is supported. When a PolarDB for MySQL instance restarts after a crash or an upgrade, data in the buffer pool is not cleared. This significantly accelerates the restart process and allows you to maintain the same level of system performance after the restart. .
  • 8.0.1.1.14(20210423)
    Category Description
    New features and optimized performance
    • The index maintenance process of PolarDB for MySQL has been optimized for a shorter response time on read-only nodes in scenarios where a large number of tables exist in the database, such as providing SaaS.
    • The synchronized interconnection among global database networks (GDNs) has been optimized to reduce the consumption of CPU resources by synchronized threads and improve the speed of CPUs with eight or less cores when they synchronize redo logs from clusters.
    • The AutoDop policy of the parameters for controlling the degree of parallelism has been optimized to avoid affecting the use of PARALLEL HINT and force_parallel_mode.
    Fixed issues
    • No wrong indexes will be used due to inaccurate statistics information incurred from the records_in_range function in the event of a large amount of range queries.
    • Inactivity of standby nodes due to small buffer pools is fixed when you restore the data of a full backup based on specified points in time. The bug of high memory usage of X-Engine metadata caches is fixed.
    • Two ACL DDL bugs of MySQL are fixed to avoid cluster unavailability due to deadlocks when DDL statements are executed.
    • No wrong result sets will be returned when parallel queries that include SQL_BUFFER_RESULT use aggregate functions, but do not execute GROUP BY statements.
  • 8.0.1.1.13.2(20210419)
    Category Description
    Fixed issues The following issue is fixed: The encryption data in the Transparent Data Encryption (TDE) table may be lost during the primary/secondary switchover. This results in encryption failures.
  • 8.0.1.1.13.1(20210408)
    Category Description
    Fixed issues The following issue is fixed: When queries are performed in parallel, slow queries are incurred because no constant filtering conditions are pushed down to the queried table to implement Block Nested-Loop Join (BNL).
  • 8.0.1.1.13(20210330)
    Category Description
    New features and optimized performance
    • Read-only nodes support the polar_use_statement_mdl_on_replica parameter. This setting prevents transactions of the read committed (RC) isolation level on read-only nodes from blocking DDL operations on the primary node. When read transactions on read-only nodes are concurrent with DDL operations on the primary node, read transactions will read different table definitions. For example, if an ADD COLUMN operation on the primary node is performed between two statements of a read transaction on the read-only node, the second statement on the read-only node reads more columns than the first statement.
    • Unnecessary equality conditions on indexed values in a query are removed. This pushes down the Limit offset statement when the ORDER BY and LIMIT clauses are used.
    • The dbms_stats.gather_stats(timeout, workers) command is added. You can update the outdated histogram by manually executing this command or using scheduled events.
    • You can query data of the log_version, log_id, origin_start_time, and rds_ext fields in the mysql.slow_log table.
    Fixed issues
    • The following issue is fixed: Queries cannot be canceled when you execute CHECK TABLE, COUNT(*), or DDL statements on X-Engine tables in an archive database.
    • KICKOUT is changed to a non-reserved keyword.
    • The following issue is fixed: Parallel queries are not performed as expected when the evaluated number of rows to be scanned in the execution plan is less than the actual number of scanned rows.
  • 8.0.1.1.12.2(20210312)
    Category Description
    Fixed issues
    • Fix the issue that the cluster becomes unavailable when you create or delete temporary tables during the storage process. This issue occurs when the session_track_temporary_tables system variable is enabled.
    • The patch for MySQL 8.0.14 is provided to fix the issue that CREATE USER statements cannot be executed as normal because the metadata locks (MDLs) of system tables in the MySQL database cannot be obtained.
  • 8.0.1.1.12.1(20210302)
    Category Description
    New features and optimized performance The write performance of X-Engine in PolarDB Archive Database is optimized when you continuously import data to multiple tables.
    Fixed issues The following issue is fixed: The state of the metadata lock (MDL) of the leader thread may be different from that of the worker thread in parallel queries because the leader thread does not own mutual exclusion locks (mutexes).
  • 8.0.1.1.12(20210220)
    Category Description
    New features and optimized performance
    • The auto_dop_low_degree_cost parameter is supported in the policies for the degree of parallelism. This parameter is used to determine how to automatically specify the degree of parallelism for parallel queries. For more information, see DOP policy.
    • The restore_table command is supported to allow you to quickly restore tables from the recycle bin. For more information, see Recycle bin.
    • You are allowed to retrieve binary logs from read-only nodes. For more information, see Remotely obtain and parse binary logs from a PolarDB for MySQL cluster.
    • The key information such as in_memory can be printed in opt trace. This allows you to quickly locate causes when an exception occurs to an execution plan.
    Fixed issues
    • The Port Percona patch is released to accelerate the detection of ACL Cache MDL conflicts in high concurrency scenarios.
    • The preferred_ordering_index parameter is added for the optimizer_switch system variable to fix the issue that the optimal plan is not used due to ordered indexes in some scenarios. This issue occurs when you use the LIMIT clause in the ORDER BY statement or the GROUP BY statement.
    • The following issue is fixed: Incorrect result may be returned when you run the SHOW PROCESSLIST command in some scenarios.
    • The following issue is fixed: For PolarDB for MySQL 8.0.1.1.10 or earlier, after you upgrade the minor version, results cannot be quickly returned when you query system tables. This is because the definition of the information_schame.KEY_COLUMN_USAGE system table is not updated.
  • 8.0.1.1.11(20210129)
    Category Description
    New features and optimized performance
    • The parallel_degree_policy parameter is added to set strategies for the degree of parallelism in parallel queries. For more information, see DOP policy.
    • The SET GLOBAL statement is supported to specify the max_digest_length parameter. This parameter is used to limit the length of recognizable statements.
      Note After you change the value of max_digest_length, the client must reconnect to the cluster for the new value to take effect.
    Fixed issues
    • The following issue is fixed: Permissions on the primary node and read-only nodes are different.
    • The following issue is fixed: Read-only nodes cannot connect to the primary node after a failover.
    • The following issue is fixed: The processing logic of the SPM is incorrect when a specific execution plan becomes invalid.
  • 8.0.1.1.10(20210112)
    Category Description
    New features and optimized performance
    • The GROUP BY statement is supported for implicit sort. You can use this statement in the same way you use it in PolarDB for MySQL 5.7.
    • Parallel queries can be disabled if a BLOB field exists.
    • The cache information about statement concurrency control can be automatically updated on read-only nodes.
    • The hot row optimization feature is supported. For more information, see Hot row optimization.
    • The physical replication optimization feature of DDL operations is supported. For more information, see DDL physical replication optimization.
    • The async metadata lock replication feature is supported. For more information, see Async Metadata Lock Replication.
    • Quick reverse traversal is supported when computations are pushed down to InnoDB.
    • The file system is optimized to speed up the process of opening multiple tables at a time.
    • The time consumed in a failover is reduced and the recovery of the new primary node is accelerated in multi-table scenarios.
    Fixed issues
    • The following issue is fixed: The system tables are lost when a read-only node is promoted to the primary node.
    • The following issue is fixed: A large number of scanned rows are evaluated when range queries are used after the parallel query feature is enabled.
    • The following issue is fixed: The results of aggregate queries are of the integral type if the field type is BIT.
    • The following issue is fixed: Incorrect results are returned when you use the SELECT DISTINCT statement after enumerated fields are used.
    • The following issue is fixed: An exception occurs in the results of parallel queries that use the EXISTS condition.
    • The following issue is fixed: Read-only nodes fail to restart in some cases.
    • The following issue is fixed: An exception occurs in table information in the data dictionary when a read-only node executes DDL statements. This is because the table for which DDL statements are being executed is re-opened when a foreign key association table is opened.
    • The following issue is fixed: The full-text index query fails due to the improper setting of a node restart flag after a failover.
    • The following issue is fixed: The log application threads on read-only nodes are blocked due to a metadata lock (MDL).
    • The following issue is fixed: The new primary node becomes unavailable after a failover because the released memory is reused.
    • The following issue is fixed: All nodes become unavailable due to the polar.info data error.
    • The following issue is fixed: An auto-increment column exception occurs in partitioned tables.
    • The following issue is fixed: A data error occurs when redo logs on the primary node are overwritten.
    • The following issue is fixed: The primary node is unavailable when it is waiting for an MDL.
    • The issues related to the transparent data encryption (TDE) feature are fixed.
    • The following issue is fixed: The cluster becomes unavailable when you execute the LOCK TABLE statement and enable the recycle bin feature.
    • The following issue is fixed: A deadlock occurs when DDL statements are executed on the primary node.
    • The following issue is fixed: The thread pool and the connection control cannot take effect at the same time.
  • 8.0.1.1.9(20201218)
    Category Description
    New features and optimized performance The SPM and PLAN keywords are removed. This prevents the issue where you cannot manage the tables whose names contain these two words.
  • 8.0.1.1.8(20201209)
    Category Description
    New features and optimized performance
    • The multi-plan mode is supported in the execution plan manager.
    • The rds_ap_threshold system variable is added to block the requests that cause the system to scan a large amount of data. These requests are evaluated by the optimizer.
    • The efficiency of writing dirty pages to disks on the primary node is improved.
    • A multi-shard writing mechanism for redo logs is supported.
    Fixed issues
    • The following issue is fixed: A null pointer of the MDL key occurs in parallel queries.
    • The following issue is fixed: A query fails when you create a parallel thread cache.
    • The following issue is fixed: An error is returned when multi-range read (MRR) is used in parallel queries.
  • 8.0.1.1.7(20201116)
    Category Description
    New features and optimized performance
    • The efficiency of scanning driven tables in parallel is improved in scenarios such as Join queries.
    • Remaining binary logs can be cleared when the binary logging feature is disabled.
    • Interrupted physical replications of slave nodes can be automatically detected and then resumed. This way, physical replications can be resumed at the earliest opportunity when they are interrupted.
    • The efficiency of the failover between the primary node and read-only nodes is improved.
    • Quick startup is supported for clusters that contain a large number of tables. This way, data files can be rapidly scanned.
    Fixed issues
    • The following issue is fixed: A cluster breakdown may occur when you obtain the type of trx->wait_lock.
    • The following issue is fixed: The number of AIO threads reaches an upper limit when a multi-queue simulated AIO is enabled.
    • The following issue is fixed: A query cannot be cancelled if a query initialization failure occurs when you query indexes.
    • The following issue is fixed: The Next Page of the current cursor points to a page that does not exist when a secondary node is in the split merge operation (SMO).
    • The following issue is fixed: Read-only nodes read logs that have been overwritten by the primary node.
    • The following issue is fixed: The system fails to clear redo logs when the timestamp intervals of redo logs are large.
    • The following issue is fixed: The table cache information in the related cache is not cleared when metadata locks are released.
  • 8.0.1.1.6(20200921)
    Category Description
    New features and optimized performance
    • The compatibility between the SPM and parallel queries is improved.
    • The efficiency of merging and sorting parallel queries is improved.
    • DELETE operations can be pushed down to InnoDB.
    • The CTS feature of PolarDB is supported.
    Fixed issues
    • The following issue is fixed: The description of pq_optimize_switch is changed.
    • The following issue is fixed: Subqueries cannot be stably executed.
  • 8.0.1.1.5(20200819)
    Category Description
    New features and optimized performance
    • A feature is added to check whether the primary node immediately applies the checkpoint policy when it establishes a replication relationship with read-only nodes.
    • Simple range queries are supported. Range queries can be processed in InnoDB.
    • The polar file system (PFS) adds the pfs_remount feature. This feature fixes the issue that files in PFS fail to be mounted when the files are open.
    • The performance bottlenecks caused by forcible pauses of parse threads on read-only nodes are removed. The efficiency of data synchronization between the primary node and read-only nodes during physical replication is improved.
    • The performance of Early Lock Release (ELR) is improved in multi-connection scenarios. The cluster performance in multi-connection scenarios is improved by 10 times.
    Fixed issues
    • The following issue is fixed: Read-only nodes become unavailable after these nodes fail to connect to the primary node.
    • The following issue is fixed: After a full-text index is used and DDL statements are executed, a read-only node becomes unavailable after a failover.
    • The following bug is fixed: Binary logs cannot be purged after the UNDO TRUNCATE command is run.
    • The following issue is fixed: Statistics are inconsistent between the primary node and read-only nodes.
  • 8.0.1.1.4(20200704)
    Category Description
    New features and optimized performance
    • DDL statements can be executed in parallel. This improves the efficiency of DDL statements.
    • The length of each queue for the simulated AIO can be dynamically adjusted.
    • Data in an FTS index cache can be synchronized between the primary node and read-only nodes. This ensures cache coherence.
    • The subqueries that include aggregate functions are supported in a WHERE clause. If the subqueries support index scans, these subqueries can also be executed in parallel.
    • The lock mode can be checked for temporary tables and standard tables.
    Fixed issues
    • The following issue is fixed: Instances become unavailable because some DDL statements are being replicated when the primary node serves as a read-only node.
    • The following issue is fixed: The performance is deteriorated because thread pools are enabled.
    • The following issue is fixed: Deadlocks occur because binary logs are purged.
    • Several issues that are related to a memory leak are fixed.
    • Several issues that occur in high availability scenarios are fixed.
  • 8.0.1.1.3(20200529)
    Category Description
    New features and optimized performance
    • Security is enhanced. For example, password management and other security operations are improved.
    • The performance of parallel queries is improved in the following scenarios:
      • The GROUP BY, UNION, and SELECT COUNT(*) FROM <table> statements are executed.
      • Shared InnoDB temporary tables are used for execution plans in parallel subqueries.
      • The views, derived tables, and temporary tables are used in execution plans.
      • Temporary tables can be used for parallel queries. However, take note of the following limits:
        • If you do not specify conditions in the SELECT COUNT(*) statement to query data in temporary tables, parallel queries are not supported.
        • Parallel queries are not supported in temporary tables for which the Memory Engine is used.
    • A new version of the audit log is supported. If audit logs are collected in this format, the log file includes a field that contains virtual IP addresses.
    • The rate of idle space to total space on index pages can be managed. This reduces latch contention and the probability of performing structure modification operations (SMOs), and improves write performance.
    • Simulated AIO among multiple queues is supported. This improves dirty page flushing and write performance.
    • Buffer pool pages can be excluded from core files. This reduces the sizes of the core files and minimizes the impact on online services.
    Fixed issues
    • In earlier versions, if the upper limit of memory in TempTable was reached, the TempTable storage engine reported an OOM error. In this case, the data is not moved from memory to disks. This issue is fixed in this version.
    • The following issue is fixed: Instances become unavailable when the sort buffer size is set to a small value and ORDER BY is used in InnoDB FTS.
    • The following issue is fixed: If a temporary table contains duplicate column names, the requested fields fail to be returned.
    • The following issue is fixed: In parallel queries, when the MAX or MIN function is used in conjunction with GROUP BY and a loose index scan is used, the statement is still being executed after the query is killed.
    • Several issues that are related to failovers are fixed.
    • Several issues that are related to parallel queries are fixed.
    • The following issue is fixed: If you execute the SHOW BINARY LOGS statement, transactions may not be committed.
  • 8.0.1.1.2(20200409)
    Category Description
    New features and optimized performance
    • The sorting method is optimized. Column values are first sorted based on string prefixes, and then based on entire strings. To be more specific, if the string prefixes of the column values are the same, the column values are sorted based on the entire strings. When columns of the CHAR and VARCHAR data types are sorted, you can specify the maximum length of different strings in the prefix to accelerate the comparison process. This reduces the sorting time.
    • Parallel queries are supported in the following scenarios:
      • Parallel queries are supported for cost models of range scans.
      • Parallel queries are supported for temporary tables.
      • Parallel queries can be performed based on the semi-join materialization strategies: lookup and scan.
    • Three types of session state trackers are added. The PolarDB intelligent proxy can use these session state trackers for connection maintenance. After the trackers are enabled, you can track the changes of user variables in sessions, CREATE and DELETE operations on temporary tables, and PREPARE and DEALLOCATION operations in SQL statements.
    • The performance of dropping an adaptive hash index (AHI) is optimized when DDL statements are executed. This reduces the impact of DDL statements on instance performance.
    • The recycle bin feature is added to prevent data loss that is caused by unintended deletion operations.
    • The performance of truncating temporary tablespaces in large buffer pools is improved. This reduces the impact of the operations that are related to temporary tables on instance performance.
    Fixed issues
    • The following issue is fixed: If aggregate functions are nested in the IF function, an issue occurs when the ROLLUP function is executed.
    • The following issue is fixed: Several issues that occur when the values of the BLOB data type are sorted.
    • The issue that occurs when the PREPARE statement includes aggregate functions in parallel queries is fixed.
    • Several issues that are related to parallel queries are fixed.
    • The following issue is fixed: Excessive redo logs may be cleared.
    • The issues that are related to the redo logs of read-only nodes are fixed.
  • 8.0.1.1.1(20200328)
    Category Description
    New features and optimized performance
    • Parallel queries are supported when subqueries contain the ROLLUP statement.
    • The statement concurrency control feature is supported.
    • The POLARDB_INDEX hint is added.
    • The replication delay between the primary node and read-only nodes is reduced.
    • The thread pool feature is supported. For more information, see Thread pool.
    • The keyring_rds plug-in for transparent data encryption (TDE) is supported.
    • GDNs are supported. For more information, see Global database networks.
    • The lock-free transaction system is optimized and the read and write performance is improved.
    Fixed issues
    • Several issues that are related to parallel queries are fixed.
    • The following issue is fixed: A value of 0 is returned for the statistics in online DDL operations.
    • The file system in user mode is optimized to accelerate instance startup.
    • The following issue is fixed: Instances may become unavailable if innodb_flush_method is set to all_o_direct.
    • The following issue is fixed: Instances may become unavailable if you release a lock after a transaction is committed.
    • The following issue is fixed: User requests may be blocked when slow query logs are truncated.
    • The following issue is fixed: Instances may become unavailable because of the compressed pages on read-only nodes.
    • The following issue is fixed: Replication connections may be killed in thread pools by mistake.
  • 8.0.1.1.0(20200205)
    Category Description
    New features and optimized performance
    • The parallel query feature is improved, and the parallel computing of the ROLLUP function that is used for enterprise-level analysis is supported.
    • The capabilities of the optimizer cost model are improved. You can specify filters to decrease selectivity. The cost model is supported for parallel queries. The optimizer checks whether a parallel or serial execution plan is used for the executed SQL statements based on the selectivity.
    • Parallel threads can be allocated based on the first in, first out (FIFO) mode and managed in a unified manner. This ensures that system resources are not exhausted by a large number of parallel queries.
    Fixed issues
    • Memory-related issues in parallel queries are fixed.
    • Issues that are related to instability in parallel queries are fixed.
  • 8.0.1.0.6(20200101)
    Category Description
    Fixed issues
    • The following issue is fixed: The binary log index file is not closed after the primary node serves as a read-only node.
    • The following issue is fixed: Instances become unavailable when read-only nodes access undo log pages that have been purged.
    • The following issue is fixed: Background threads access tablespace pages that do not exist if a read-only node serves as the primary node during a failover.
    • The following issue is fixed: After clusters are shut down, log threads exit but redo logs are still being written. In this case, the clusters become unavailable.
  • 8.0.1.0.5(20191203)
    Category Description
    New features and optimized performance
    • The optimizer trace allows you to trace information about parallel queries. For example, you can use the optimizer trace tool to analyze the reason why a parallel query is used or not used.
    • Hints are added for parallel queries. You can add hints to SQL statements to explicitly enable parallelism and specify the degree of parallelism.
    • Parallel scans are supported if you execute the INSERT... SELECT statement when the isolation level is READ COMMITTED. You can execute the INSERT...SELECT statement to query data and insert the result into another table.
    Fixed issues
    • Several issues that are related to parallel queries are fixed.
    • The following issue is fixed: Instances become unavailable when a read-only node is promoted to the primary node during a failover.
    • The following issue is fixed: Errors are reported due to the execution of some DDL statements when a failover occurs.
    • The following issue is fixed: An error message "too many connection error" is reported if the upper limit of connections that are supported by locks is reached.

PolarDB for MySQL 5.7 release notes

The following table describes the versions and the features that are supported by PolarDB for MySQL 5.7.

Release date Revision version MySQL minor version with full compatibility Released feature
2021-04-19 5.7.1.0.8 5.7.28 A feature is added to check whether the primary node immediately applies the checkpoint policy when it establishes a replication relationship with read-only nodes.
2021-03-10 5.7.1.0.7
  • The parallel DDL feature is supported to speed up the execution of DDL operations. For more information, see Parallel DDL.
  • The innodb_buffer_pool_in_core_file parameter is added to remove buffer pools from core files.
  • The following keywords are supported in index hints. This allows the optimizer to use or ignore the specified index in query processing.
    • GROUP_INDEX or NO_GROUP_INDEX: uses or ignores the specified indexes for index scans in GROUP BY operations.
    • INDEX or NO_INDEX: forces the server to use or ignore the specified indexes.
    • JOIN_INDEX or NO_JOIN_INDEX: forces MySQL to use or ignore the specified indexes for all access methods.
  • The following keywords are supported for the join order. This allows the optimizer to join tables in a proper order.
    • JOIN_FIXED_ORDER: forces the optimizer to join tables by using the order in which the tables appear in the FROM clause.
    • JOIN_ORDER: instructs the optimizer to join tables by using the specified table order. The 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: instructs the optimizer to join tables by using the specified table order for the first tables of the join execution plan. The hint applies to named tables. The optimizer places all other tables after named tables.
    • JOIN_SUFFIX: instruct the optimizer to join tables by using the specified table order for the last tables of the join execution plan. The hint applies to named tables. The optimizer places all other tables before named tables.
2021-02-22 5.7.1.0.6.3 Fixes a known issue.
2021-02-10 5.7.1.0.6.2 Fixes a known issue.
2021-02-02 5.7.1.0.6.1 Fixes several known issues.
2021-01-29 5.7.1.0.6
  • The statement queue feature is supported. For more information, see Statement Queue.
  • The instant add column feature is supported. For more information, see Instant ADD COLUMN.
  • The returning feature is supported. For more information, see Returning.
2020-12-31 5.7.1.0.5 Fixes several known issues.
2020-11-17 5.7.1.0.4 Fixes several known issues.
2020-10-21 5.7.1.0.3
2020-08-28 5.7.1.0.2 Fixes several known issues.
2020-07-30 5.7.1.0.1 Fixes several known issues.
  • 5.7.1.0.8(20210419)
    Category Description
    New features and optimized performance A feature is added to check whether the primary node immediately applies the checkpoint policy when it establishes a replication relationship with read-only nodes.
    Fixed issues
    • KICKOUT is changed to a non-reserved keyword.
    • The following issue is fixed: The encryption data in the Transparent Data Encryption (TDE) table may be lost during the primary/secondary switchover. This results in encryption failures.
  • 5.7.1.0.7 (20210310)
    Category Description
    New features and optimized performance
    • The parallel DDL feature is supported to speed up the execution of DDL operations. For more information, see Parallel DDL.
    • The innodb_buffer_pool_in_core_file parameter is added to remove buffer pools from core files.
    • The following keywords are supported in index hints. This allows the optimizer to use or ignore the specified index in query processing.
      • GROUP_INDEX or NO_GROUP_INDEX: uses or ignores the specified indexes for index scans in GROUP BY operations.
      • INDEX or NO_INDEX: forces the server to use or ignore the specified indexes.
      • JOIN_INDEX or NO_JOIN_INDEX: forces MySQL to use or ignore the specified indexes for all access methods.
    • The following keywords are supported for the join order. This allows the optimizer to join tables in a proper order.
      • JOIN_FIXED_ORDER: forces the optimizer to join tables by using the order in which the tables appear in the FROM clause.
      • JOIN_ORDER: instructs the optimizer to join tables by using the specified table order. The 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: instructs the optimizer to join tables by using the specified table order for the first tables of the join execution plan. The hint applies to named tables. The optimizer places all other tables after named tables.
      • JOIN_SUFFIX: instruct the optimizer to join tables by using the specified table order for the last tables of the join execution plan. The hint applies to named tables. The optimizer places all other tables before named tables.
    Fixed issues
    • The mapping of token number 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.
    • Fix the issue that a configuration change may fail because read-only nodes cannot be registered on the primary node.
    • Fix the issue that the cluster becomes unavailable when you create or delete temporary tables during the storage process. This issue occurs when the session_track_temporary_tables system variable is enabled.
  • 5.7.1.0.6.3 (20210222)
    Category Description
    Fixed issues The following issue is fixed: An incorrect result is returned after you execute the SHOW PROCESSLIST statement in some scenarios.
  • 5.7.1.0.6.2 (20210210)
    Category Description
    Fixed issues Fixes the issue that an error occasionally occurs when you perform a query that is optimized by Index Merge on a read-only node. The error code is 1032 and the error message is Can't find record in TABLE.
  • 5.7.1.0.6.1 (20210202)
    Category Description
    Fixed issues
    • Fixes the issue that dirty page flushing cannot be performed as expected on read and write nodes.
    • Fixed the issue that the primary zone cannot be changed on clusters that have performed a failover.
    • Fixed the issue that the database crashes if Thread Descriptor (THD) is accidentally invoked when SHOW PROCESSLIST is executed.
  • 5.7.1.0.6 (20210129)
    Category Description
    New features and optimized performance
    • The statement queue feature is supported. For more information, see Statement Queue.
    • The instant add column feature is supported. For more information, see Instant ADD COLUMN.
    • The returning feature is supported. For more information, see Returning.
    Fixed issues Fixed the issue that the system table is lost when you upgrade the PolarDB kernel version.
  • 5.7.1.0.5 (20201231)
    Category Description
    Fixed issues
    • Fixes the issue that read-only nodes crash when you query data from read-only nodes after the key of the primary node is modified.
    • Fixes the issue that read-only nodes crash when you analyze partition tables on read-only nodes.
    • Fixes the issue that the new primary node crashes due to tablespace inconsistency after a failover.
  • 5.7.1.0.4 (20201117)
    Category Description
    Fixed issues
    • Fixes the issue that data cannot be inserted into temporary tables after a failover.
    • Fixes the issue that a read-only node is unavailable when a temporary tablespace is extended on the read-only node.
    • Fixes the issue that the simulated AIO does not work as expected.
  • 5.7.1.0.3 (20201021)
    Category Description
    New features and optimized performance Supports TDE.
    Fixed issues
    • Fixes the issue that statistics are inconsistent between the primary node and read-only nodes.
    • Fixes the issue that an error is returned when you execute the SHOW PROCESSLIST statement.
  • 5.7.1.0.2 (20200828)
    Category Description
    Fixed issues
    • Fixes the issue that .IDB files cannot be created after a migration from ApsaraDB RDS for MySQL 5.7.
    • You are not allowed to run CF_STATUS_COMMAND commands during a failover process. This ensures the success of the failover.
    • Fixes the issue that the primary node is unavailable due to the page contention that is caused by background updates for statistics thread and the truncate operation.
  • 5.7.1.0.1 (20200730)
    Category Description
    Fixed issues
    • Fixes the issue that maxscale is unavailable due to the missing code of specific PolarDB dedicated commands.
    • Fixes the issue that the cluster that you create from the cluster recycle bin is unavailable because no tablespace can be found.
    • Fixes the issue that the unmout PFS file crashes if one or more files are still open when you run the promote replica command.

PolarDB for MySQL 5.6 release notes

The following table describes the versions and features that are supported by PolarDB for MySQL 5.6.

Version Release date MySQL minor version with full compatibility Released feature
5.6.1.0.26 2021-03-19 5.6.16 Fixes several known issues.
5.6.1.0.25 2021-02-05 The database and table restoration is optimized and sped up.
5.6.1.0.24 2021-01-22
  • The initialization of the PolarDB engine is optimized and the time consumed to start up the engine is reduced in large table scenarios.
  • You can view the Kernel Version of a cluster on the Basic Information page.
5.6.1.0.23 2021-01-04 The following issue is fixed: A memory leak occurs on read-only nodes.
5.6.1.0.22 2020-12-25 Directory indexes are supported in polar file system (PFS). This improves the performance of the cluster that has a large number of tables.
5.6.1.0.21 2020-11-12 The transparent data encryption (TDE) feature and data restoration of specified databases and tables are supported. For more information, see TDE and Restore data of specific databases and tables.
5.6.1.0.20 2020-10-27 The efficiency of physical replications in some scenarios is improved.
20200831 2020-09-22 PFS supports local disks and writable snapshots. The performance of PFS is optimized.
20200616 2020-07-01 The statement queue feature is supported. For more information, see Statement Queue.
20200601 2020-06-05 The statement concurrency control feature is supported to improve concurrency performance. For more information, see Statement Concurrency Control.
20200507 2020-05-13 The performance bottlenecks that are caused by internal data structures are removed. The simulated I/O system is optimized.
  • 5.6 (5.6.1.0.26)
    Category Description
    Fixed issues
    • The following issue is fixed: The connection may fail when you run the FLUSH PRIVILEGES or FLUSH GRANT command.
    • The following issue is fixed: An incorrect estimation result is returned in some cases because the estimation of partition tables is aborted.
    • The following issue is fixed: An incorrect result is returned after you execute the SHOW PROCESSLIST statement in some scenarios.
    • The following issue is fixed: The cluster becomes unavailable when you create or delete temporary tables during the storage process. This issue occurs when the session_track_temporary_tables system variable is enabled.
  • 5.6 (5.6.1.0.25)
    Category Description
    New features and optimized performance The database and table restoration is optimized and sped up.
    Fixed issues
    • The following issue is fixed: Read-only nodes may be unavailable after they read truncated undo pages.
    • The following issue is fixed: The primary zone cannot be changed on clusters that have performed a failover.
  • 5.6 (5.6.1.0.24)
    Category Description
    New features and optimized performance
    • The initialization of the PolarDB engine is optimized and the time consumed to start up the engine is reduced in large table scenarios.
    • You can view the Kernel Version of a cluster on the Basic Information page.
    Fixed issues
    • The following issue is fixed: Undo logs cannot be truncated when data is migrated from an ApsaraDB RDS instance to PolarDB.
    • The following issue is fixed: System tables cannot be added.
    • The following issue is fixed: The primary node is unavailable when databases and tables are restored.
    • The following issue is fixed: The result is not correctly sorted when a query returns a DECIMAL result.
    • Several issues are fixed: The MySQL service process may crash in specific scenarios.
  • 5.6 (5.6.1.0.23)
    Category Description
    Fixed issues The following issue is fixed: A memory leak occurs on read-only nodes.
  • 5.6 (5.6.1.0.22)
    Category Description
    New features and optimized performance Directory indexes are supported in polar file system (PFS). This improves the performance of the cluster that has a large number of tables.
    Fixed issues
    • The following issue is fixed: The cluster becomes unavailable after a failover between the primary node and read-only nodes due to invalid node roles in specific cases.
    • The following issue is fixed: Read-only nodes do not response due to the uninitialized statement queue.
    • The following issue is fixed: New system tables are not initialized after a failover between the primary node and read-only nodes.
    • The following issue is fixed: The thread pool and connection control features are enabled at the same time.
    • The following issue is fixed: A cluster becomes unavailable due to duplicate IDs in the full-text index.
    • The following issue is fixed: You cannot query data from read-only nodes in some cases.
    • The following issue is fixed: The replication is interrupted when the log replication thread accidentally exits.
  • 5.6 (5.6.1.0.21)
    Category Description
    New features and optimized performance
    • The data restoration of specified databases and tables are supported. For more information, see Restore data of specific databases and tables.
    • Supports TDE.
    • PFS allows you to mount multiple disks to restore individual tables.
    • Resource occupation that is caused by the Adaptive Hash Index (AHI) lock during physical replications is optimized.
    Fixed issues
    • The following issue is fixed: The out of memory (OOM) error occurs when you use DYNAMIC RANGE and INDEX MERGE in SELECT statements.
    • The following issue is fixed: A cluster breaks down when you create or delete an account in some cases.
    • The following issue is fixed: Standby nodes cannot be reconnected in some cases.
    • The following issue is fixed: The cluster fails to start when a failover error occurs.
    • The following issue is fixed: The state of the Binlog thread is not as expected in some cases.
  • 5.6 (5.6.1.0.20)
    Category Description
    New features and optimized performance The efficiency of physical replications in some scenarios is improved.
    Fixed issues
    • The following issue is fixed: A cluster breaks down when you execute the CREATE TABLE... SELECT statement in some cases.
    • The following issue is fixed: A memory leak occurs in the stored procedure due to the excessive use of derived tables.
    • The following issue is fixed: The restore time is invalid or the restoration fails. This issue occurs when you restore data to a specified point in time.
    • The following issue is fixed: A large number of logs about PFS exceptions are reported.
    • The following issue is fixed: Tables are lost when you perform DDL operations. This issue occurs after foreign key check is disabled.
    • The following issue is fixed: Read-only nodes break down when you truncate multiple temporary tables at a time.
  • 5.6(20200831)
    Category Description
    New features and optimized performance PFS supports local disks and writable snapshots. The performance of PFS is optimized.
    Fixed issues
    • The following issue is fixed: The cluster breaks down due to the uninitialized statement queue.
    • The following issue is fixed: A large amount of storage is consumed by the corefile.
    • The following issue is fixed: Statistics are inconsistent between the primary node and read-only nodes.
    • The following issue is fixed: After a read-only node is specified as the primary node, the other read-only nodes cannot connect to the new primary node.
    • The following issue is fixed: The full-text index caches are inconsistent between the primary node and read-only nodes.
  • 5.6(20200616)
    Category Description
    New features and optimized performance
    • The statement queue feature is supported. For more information, see Statement Queue.
    • The replication delay between the primary node and read-only nodes is reduced.
    • The metadata lock can be replicated when you execute the LOCK TABLE statement.
    Fixed issues
    • The metadata lock issue that occurs during data replication is fixed.
    • The following issue is fixed: A full-text search (FTS) index is created in the system table.
    • The issues that are related to the updates and performance optimization of hot rows are fixed.
  • 5.6(20200601)
    Category Description
    New features and optimized performance
    Fixed issues The following issue is fixed: The write performance is reduced due to the issues of thread pools.
  • 5.6(20200507)
    Category Description
    New features and optimized performance
    • The statement concurrency control feature is added.
    • A parameter is added to manage the ratio of idle space to total space on index pages.
    • The simulated AIO model is optimized.
    Fixed issues
    • The following issue is fixed: The performance is reduced due to the BoolFlag data type.
    • The following issue is fixed: Clusters become unavailable when the pfs_umount table is not closed.

Upgrade versions

  • Upgrade DB versions and minor versions

    PolarDB does not allow you to upgrade the DB versions or the minor versions in a direct way. For example, you cannot upgrade PolarDB for MySQL 5.6 to PolarDB for MySQL 8.0 or upgrade the minor version of from PolarDB for MySQL 8.0.1 to 8.0.2 in a direct way. However, you can use Data Transmission Service (DTS) to migrate or synchronize data from the cluster of the source version to the cluster of the destination version. For more information about how to migrate or synchronize data, see Overview of data migration or synchronization solutions.

  • Upgrade revision versions

    For more information about how to upgrade revision versions, see Upgrade the revision version.