Aurora MySQL database engine updates 2020-09-17 (version 2.09.0) - Amazon Aurora

Aurora MySQL database engine updates 2020-09-17 (version 2.09.0)

Version: 2.09.0

Aurora MySQL 2.09.0 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 1.14.*, 1.15.*, 1.16.*, 1.17.*, 1.18.*, 1.19.*, 1.20.*, 1.21.*, 1.22.*, 1.23.*, 2.01.*, 2.02.*, 2.03.*, 2.04.*, 2.05.*, 2.06.*, 2.07.*, 2.08.*, and 2.09.*.

You can restore a snapshot from a currently supported Aurora MySQL release into Aurora MySQL 2.09.0. You also have the option to upgrade existing Aurora MySQL 2.* database clusters to Aurora MySQL 2.09.0. You can't upgrade an existing Aurora MySQL 1.* cluster directly to 2.09.0; however, you can restore its snapshot to Aurora MySQL 2.09.0.

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

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

Note

For information on how to upgrade your Aurora MySQL database cluster, see Database upgrades and patches for Amazon Aurora MySQL.

Improvements

New features:

  • With this release, you can create Amazon Aurora MySQL database instances with up to 128 tebibytes (TiB) of storage. The new storage limit is an increase from the prior 64 TiB. The 128 TiB storage size supports larger databases. This capability is not supported on small instances sizes (db.t2 or db.t3). A single tablespace cannot grow beyond 64 TiB due to InnoDB limitations with 16 KB page size.

    Aurora alerts you when the cluster volume size is near 128 TiB, so that you can take action prior to hitting the size limit. The alerts appear in the mysql log and RDS Events in the AWS Management Console.

  • You can now turn parallel query on or off for an existing cluster by changing the value of the DB cluster parameter aurora_parallel_query. You don't need to use the parallelquery setting for the --engine-mode parameter when creating the cluster.

    Parallel query is now expanded to be available in all regions where Aurora MySQL is available.

    There are a number of other functionality enhancements and changes to the procedures for upgrading and enabling parallel query in an Aurora cluster. For more information, see Working with parallel query for Amazon Aurora MySQL.

  • Aurora dynamically resizes your cluster storage space. With dynamic resizing, the storage space for your Aurora DB cluster automatically decreases when you remove data from the DB cluster. For more information, see Storage scaling.

    Note

    The dynamic resizing feature is being deployed in phases to the AWS Regions where Aurora is available. Depending on the Region where your cluster is, this feature might not be available yet. For more information, see the What's New announcement.

High priority fixes:

  • Backport of Community Bug #27659490: SELECT USING DYNAMIC RANGE AND INDEX MERGE USE TOO MUCH MEMORY (OOM)

  • Bug #26881508: MYSQL #1: DISABLE_ABORT_ON_ERROR IN AUTH_COMMON.H

  • Backport of Community Bug #24437124: POSSIBLE BUFFER OVERFLOW ON CREATE TABLE

  • Backport of Bug #27158030: INNODB ONLINE ALTER CRASHES WITH CONCURRENT DML

  • Bug #29770705: SERVER CRASHED WHILE EXECUTING SELECT WITH SPECIFIC WHERE CLAUSE

  • Backport of BUG #26502135: MYSQLD SEGFAULTS IN MDL_CONTEXT::TRY_ACQUIRE_LOCK_IMPL

  • Backport of Bug #26935001: ALTER TABLE AUTO_INCREMENT TRIES TO READ INDEX FROM DISCARDED TABLESPACE

  • Bug #28491099: [FATAL] MEMORY BLOCK IS INVALID | INNODB: ASSERTION FAILURE: UT0UT.CC:670

  • Bug #30499288: GCC 9.2.1 REPORTS A NEW WARNING FOR OS_FILE_GET_PARENT_DIR

  • Bug #29952565 where MYSQLD GOT SIGNAL 11 WHILE EXECUTING A QUERY(UNION + ORDER BY + SUB-QUERY)

  • Bug #30628268: OUT OF MEMORY CRASH

  • Bug #30441969: BUG #29723340: MYSQL SERVER CRASH AFTER SQL QUERY WITH DATA ?AST

  • Bug #30569003: 5.7 REPLICATION BREAKAGE WITH SYNTAX ERROR WITH GRANT MANAGEMENT

  • Bug #29915479: RUNNING COM_REGISTER_SLAVE WITHOUT COM_BINLOG_DUMP CAN RESULTS IN SERVER EXIT

  • Bug #30569003: 5.7 REPLICATION BREAKAGE WITH SYNTAX ERROR WITH GRANT MANAGEMENT

  • Bug #29915479: RUNNING COM_REGISTER_SLAVE WITHOUT COM_BINLOG_DUMP CAN RESULTS IN SERVER EXIT

  • Bug #20712046: SHOW PROCESSLIST AND PERFORMANCE_SCHEMA TABLES DO NOT MASK PASSWORD FROM QUERY

  • Backport bug #18898433: EXTREMELY SLOW PERFORMANCE WITH OUTER JOINS AND JOIN BUFFER (fixed in 5.7.21). Queries with many left joins were slow if join buffering was used (for example, using the block nested loop algorithm). (Bug #18898433, Bug #72854)"

  • Backport bug #26402045: MYSQLD CRASHES ON QUERY (fixed in MySQL 5.7.23). Certain cases of subquery materialization could cause a server exit. These queries now produce an error suggesting that materialization be disabled. (Bug #26402045)

  • [Backport from MySQL] users other than rdsadmin is disallowed to update pfs table in the reader replica.

  • Fix the issue where the customer can not update the perfschema in the reader replica

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

  • Bug #26997096: relay_log_space value is not updated in a synchronized manner so that its value sometimes much higher than the actual disk space used by relay logs.

  • BUG #25082593: FOREIGN KEY VALIDATION DOESN'T NEED TO ACQUIRE GAP LOCK IN READ COMMITTED

  • CVE-2019-2731

  • CVE-2018-2645

  • CVE-2019-2581

  • CVE-2018-2787

  • CVE-2019-2482

  • CVE-2018-2640

  • CVE-2018-2784

  • CVE-2019-2628

  • CVE-2019-2911

  • CVE-2019-2628

  • CVE-2018-3284

  • CVE-2018-3065

  • CVE-2019-2537

  • CVE-2019-2948

  • CVE-2019-2434

  • CVE-2019-2420

Availability improvements:

  • Enable lock manager ABA fix by default.

  • Fixed an issue in the lock manager where a race condition can cause a lock to be shared by two transactions, causing the database to restart.

  • Fixed an issue when creating a temporary table with compressed row format might result in a restart.

  • Fix default value of table_open_cache on 16XL and 24XL instances which could cause repeated failovers and high CPU utilization on large instances classes (R4/R5-16XL, R5-12XL, R5-24XL). This impacted 2.07.x.

  • Fixed an issue where restoring a cluster from Amazon S3 to Aurora MySQL version 2.08.0 takes longer than expected when the S3 backup didn't include the mysql.host table.

  • Fixed an issue which might cause repeated failovers due to updates of virtual columns with secondary indexes.

  • Fixed an issue related to transaction lock memory management with long-running write transactions resulting in a database restart.

  • Fixed multiple issues where the the engine might crash during zero-downtime patching while checking for safe point for patching.

  • Fixed an issue to skip redo logging for temporary tables, which was previously causing a crash.

  • Fixed a race condition in the lock manager between killing connection/query and the session killed.

  • Fixed an issue where the database could crash if it is a binlog replica and receives a DDL event over the MySQL time_zone table.

Global databases:

  • MySQL INFORMATION_SCHEMA.REPLICA_HOST_STATUS view in a secondary Region now shows the entries for the replicas belonging to that Region.

  • Fixed unexpected query failures that could occur in a Global DB secondary Region after temporary network connectivity issues between the primary and secondary Regions.

Parallel query:

  • Fixed the EXPLAIN plan for a Parallel Query query, which is incorrect for a simple single-table query.

  • Fixed self-deadlatch that may occur when Parallel Query is enabled.

General improvements:

  • Export to S3 now supports the ENCRYPTION keyword.

  • The aurora_binlog_replication_max_yield_seconds parameter now has a max value of 36,000. The previous maximum accepted value was 45. This parameter works only when the parameter aurora_binlog_use_large_read_buffer is set to 1.

  • Changed the behavior to map MIXED binlog_format to ROW instead of STATEMENT when executing LOAD DATA FROM INFILE | S3.

  • Fixed an issue where a binlog replica connected to an Aurora MySQL binlog primary might show incomplete data when the primary executed LOAD DATA FROM S3 and binlog_format is set to STATEMENT.

  • Increased maximum allowable length for audit system variables server_audit_incl_users and server_audit_excl_users from 1024 bytes to 2000 bytes.

  • Fixed an issue where users may lose access to the database when lowering the max_connections parameter in the parameter group when the current connections is greater than the value being set.

  • Fixed an issue in Data Activity Streams where a single quote and backslash were not escaped properly.

Integration of MySQL community edition bug fixes

  • Bug #27659490: SELECT USING DYNAMIC RANGE AND INDEX MERGE USE TOO MUCH MEMORY(OOM)

  • Bug #26881508: MYSQL #1: DISABLE_ABORT_ON_ERROR IN AUTH_COMMON.H

  • Bug #24437124: POSSIBLE BUFFER OVERFLOW ON CREATE TABLE

  • Bug #27158030: INNODB ONLINE ALTER CRASHES WITH CONCURRENT DML

  • Bug #29770705: SERVER CRASHED WHILE EXECUTING SELECT WITH SPECIFIC WHERE CLAUSE

  • Bug #26502135: MYSQLD SEGFAULTS IN MDL_CONTEXT::TRY_ACQUIRE_LOCK_IMPL

  • Bug #26935001: ALTER TABLE AUTO_INCREMENT TRIES TO READ INDEX FROM DISCARDED TABLESPACE

  • Bug #28491099: [FATAL] MEMORY BLOCK IS INVALID | INNODB: ASSERTION FAILURE: UT0UT.CC:670

  • Bug #30499288: GCC 9.2.1 REPORTS A NEW WARNING FOR OS_FILE_GET_PARENT_DIR

  • Bug #29952565: where MYSQLD GOT SIGNAL 11 WHILE EXECUTING A QUERY(UNION + ORDER BY + SUB-QUERY)

  • Bug #30628268: OUT OF MEMORY CRASH

  • Bug #30441969: BUG #29723340: MYSQL SERVER CRASH AFTER SQL QUERY WITH DATA ?AST

  • Bug #30569003: 5.7 REPLICATION BREAKAGE WITH SYNTAX ERROR WITH GRANT MANAGEMENT

  • Bug #29915479: RUNNING COM_REGISTER_SLAVE WITHOUT COM_BINLOG_DUMP CAN RESULTS IN SERVER EXIT

  • Bug #30569003: 5.7 REPLICATION BREAKAGE WITH SYNTAX ERROR WITH GRANT MANAGEMENT

  • Bug #29915479: RUNNING COM_REGISTER_SLAVE WITHOUT COM_BINLOG_DUMP CAN RESULTS IN SERVER EXIT

  • Bug #20712046: SHOW PROCESSLIST AND PERFORMANCE_SCHEMA TABLES DO NOT MASK PASSWORD FROM QUERY

  • Bug #18898433: EXTREMELY SLOW PERFORMANCE WITH OUTER JOINS AND JOIN BUFFER (fixed in 5.7.21)

  • Bug #26402045: MYSQLD CRASHES ON QUERY (fixed in MySQL 5.7.23)

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

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

  • Bug #26997096: relay_log_space value is not updated in a synchronized manner so that its value sometimes much higher than the actual disk space used by relay logs. (https://github.com/mysql/mysql-server/commit/78f25d2809ad457e81f90342239c9bc32a36cdfa)

  • Bug #25082593: FOREIGN KEY VALIDATION DOESN'T NEED TO ACQUIRE GAP LOCK IN READ COMMITTED

  • Bug #24764800: REPLICATION FAILING ON SLAVE WITH XAER_RMFAIL ERROR.

  • Bug #81441: WARNING ABOUT LOCALHOST WHEN USING SKIP-NAME-RESOLVE.