Aurora MySQL database engine updates 2020-08-14 (version 2.04.9) (Deprecated) - Amazon Aurora
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Aurora MySQL database engine updates 2020-08-14 (version 2.04.9) (Deprecated)

Version: 2.04.9

Aurora MySQL 2.04.9 is generally available. Aurora MySQL 2.x versions are compatible with MySQL 5.7 and Aurora MySQL 1.x versions are compatible with MySQL 5.6.

Currently supported Aurora MySQL releases are 1.19.5, 1.19.6, 1.22.*, 1.23.*, 2.04.*, 2.07.*, 2.08.*, 2.09.*, 2.10.*, 3.01.* and 3.02.*.

To create a cluster with an older version of Aurora MySQL, please specify the engine version through the Amazon Web Services Management Console, the Amazon CLI, or the RDS API.

If you have any questions or concerns, Amazon Support is available on the community forums and through Amazon Support. For more information, see Maintaining an Amazon Aurora DB cluster in the Amazon Aurora User Guide.

Note

This version is currently not available in the following Amazon Regions: Amazon GovCloud (US-East) [us-gov-east-1], Amazon GovCloud (US-West) [us-gov-west-1], Asia Pacific (Hong Kong) [ap-east-1], and Middle East (Bahrain) [me-south-1]. There will be a separate announcement once it is made available.

Note

For information on how to upgrade your Aurora MySQL database cluster, see Upgrading the minor version or patch level of an Aurora MySQL DB cluster in the Amazon Aurora User Guide.

Improvements

High-priority fixes:

Critical fixes:

Availability improvements:

  • Fixed an issue that could cause a database restart or failover due to execution of a kill session command. If you encounter this issue, contact Amazon support to enable this fix on your instance.

  • Fixed an issue that causes a database restart during execution of a complex query involving multi-table joins and aggregation that use intermediate tables internally.

  • Fixed an issue that causes database restarts due to an interrupted DROP TABLE on multiple tables.

  • Fixed an issue that causes a database failover during database recovery.

  • Fixed a database restart caused by incorrect reporting of threads_running when audit and slow query logs are enabled.

  • Fixed an issue where a kill query command might get stuck during execution.

  • Fixed a race condition in the lock manager that resulted in a database restart or failover during transaction rollback.

  • Fixed an issue that triggered database restart or failover when multiple connections are trying to update the same table with a Full-Text Search index.

  • Fixed an issue that can cause a deadlatch when purging an index resulting in a failover or restart.

General improvements:

  • Fixed issues that could cause queries on read replicas to use data from an uncommitted transaction. This issue is limited to the transactions that are started immediately after a database restart.

  • Fixed an issue encountered during INPLACE ALTER TABLE for a table with triggers defined and when the DDL did not contain a RENAME clause.

  • Fixed an issue that caused cloning to take longer on some database clusters with high writeload.

  • Fixed an issue encountered during an upgrade when a partitioned table has embedded spaces in the name.

  • Fixed an issue where the read replica might transiently see partial results of a recently committed transaction on the writer.

  • Fixed an issue where queries on a read replica against an FTS table may produce stale results. This will only occur when the FTS query on the read replica closely follows a query on INFORMATION_SCHEMA.INNODB_SYS_TABLES for the same FTS table on the writer.

  • Fixed an issue that resulted in a slow restore of Aurora 1.x database cluster containing FTS (Full-Text Search) indexes to an Aurora 2.x database cluster.

  • Extended maximum allowable length to 2000 for server_audit_incl_users and server_audit_excl_users global parameters.

  • Fixed an issue where Aurora 1.x to Aurora 2.x restore might take an extended time to complete.

  • Fixed an issue where a lambda_async invocation through stored procedure doesn't work with Unicode.

  • Fixed an issue encountered when a spatial index does not properly handle an off-record geometry column.

  • Fixed an issue that might cause a query to fail on a reader DB instance with InternalFailureException error with message"Operation terminated (internal error)".

Integration of MySQL bug fixes

  • Bug #23070734, Bug #80060: Concurrent TRUNCATE TABLEs cause stalls

  • Bug #23103937: PS_TRUNCATE_ALL_TABLES() DOES NOT WORK IN SUPER_READ_ONLY MODE

  • Bug#22551677: When taking the server offline, a race condition within the Performance Schema could lead to a server exit.

  • Bug #27082268: Invalid FTS sync synchronization.

  • BUG #12589870: Fixed an issues which causes a restart with multi-query statement when query cache is enabled.

  • Bug #26402045: Certain cases of subquery materialization could cause a server exit. These queries now produce an error suggesting that materialization be disabled.

  • Bug #18898433: Queries with many left joins were slow if join buffering was used (for example, using the block nested loop algorithm).

  • Bug #25222337: A NULL virtual column field name in a virtual index caused a server exit during a field name comparison that occurs while populating virtual columns affected by a foreign key constraint. (https://github.com/mysql/mysql-server/commit/273d5c9d7072c63b6c47dbef6963d7dc491d5131)

  • Bug #25053286: Executing a stored procedure containing a query that accessed a view could allocate memory that was not freed until the session ended. (https://github.com/mysql/mysql-server/commit/d7b37d4d141a95f577916448650c429f0d6e193d)

  • Bug #25586773: Executing a stored procedure containing a statement that created a table from the contents of certain SELECT (https://dev.mysql.com/doc/refman/5.7/en/select.html) statements could result in a memory leak. (https://github.com/mysql/mysql-server/commit/88301e5adab65f6750f66af284be410c4369d0c1)

  • Bug #26666274: INFINITE LOOP IN PERFORMANCE SCHEMA BUFFER CONTAINER.

  • Bug #23550835, Bug #23298025, Bug #81464: A SELECT Performance Schema tables when an internal buffer was full could cause a server exit.

Comparison with Aurora MySQL version 1

The following Amazon Aurora MySQL features are supported in Aurora MySQL Version 1 (compatible with MySQL 5.6), but these features are currently not supported in Aurora MySQL Version 2 (compatible with MySQL 5.7).

MySQL 5.7 compatibility

Aurora MySQL 2.04.9 is wire-compatible with MySQL 5.7 and includes features such as JSON support, spatial indexes, and generated columns. Aurora MySQL uses a native implementation of spatial indexing using z-order curves to deliver >20x better write performance and >10x better read performance than MySQL 5.7 for spatial datasets.

Aurora MySQL 2.04.9 does not currently support the following MySQL 5.7 features:

  • Group replication plugin

  • Increased page size

  • InnoDB buffer pool loading at startup

  • InnoDB full-text parser plugin

  • Multisource replication

  • Online buffer pool resizing

  • Password validation plugin

  • Query rewrite plugins

  • Replication filtering

  • The CREATE TABLESPACE SQL statement