Aurora MySQL database engine updates 2025-11-20 (version 3.10.2, compatible with MySQL 8.0.42)
Version: 3.10.2
Aurora MySQL 3.10.2 is generally available. Aurora MySQL 3.10 versions are compatible with MySQL 8.0.42. For more information on the community changes that have occurred, see MySQL 8.0 Release Notes
For details of the new features in Aurora MySQL version 3, see Aurora MySQL version 3 compatible with MySQL 8.0. For differences between Aurora MySQL version 3 and Aurora MySQL version 2, see Comparison of Aurora MySQL version 2 and Aurora MySQL version 3. For a comparison of Aurora MySQL version 3 and MySQL 8.0 Community Edition, see Comparison of Aurora MySQL version 3 and MySQL 8.0 Community Edition in the Amazon Aurora User Guide.
You can perform an in-place upgrade leveraging Zero Downtime Patching (ZDP), restore a snapshot, or initiate a managed blue/green upgrade using Amazon RDS Blue/Green Deployments from any currently supported Aurora MySQL version 2 cluster into an Aurora MySQL version 3.10.2 cluster.
For information on planning an upgrade to Aurora MySQL version 3, see Planning a major version upgrade for an Aurora MySQL cluster. For general information about Aurora MySQL upgrades, see Upgrading Aurora MySQL DB clusters in the Amazon Aurora User Guide.
For troubleshooting information, see Troubleshooting for Aurora MySQL in-place upgrade in the Amazon Aurora User Guide.
If you have any questions or concerns, Amazon Web Services Support is available on the community forums and through Amazon Web Services Support
Improvements
Availability improvements:
-
Fixed an issue which can cause engine restart during Zero-downtime patching (ZDP)/Zero-downtime restart (ZDR), when preserving SSL/TLS connections with active transactions.
-
Fixed an issue which, can cause an engine restart when running
KILL <query-id>after runningEXPLAIN FOR CONNECTION <query-id>on a running Parallel Query. -
Fixed an issue which could cause writer to become unresponsive when write forwarding is being disabled while two or more replicas are forwarding requests.
-
Fixed an issue which could cause the writer instance to restart when global write forwarding or local write forwarding is being disabled.
General improvements:
-
Fixed an issue causing inaccurate tracking of Parallel Query requests while running
EXPLAIN ANALYZEstatements whereAurora_pq_request_in_progresscounter was not updated accurately. -
Automatically disable
aurora_oom_responseactions (except print, if configured) whenaurora_oom_responsefails to resolve memory pressure after a threshold time (in the order of few mins). -
Improved write IOPS performance when system variable
innodb_flush_log_at_trx_commitis set to0. -
Fixed an issue with replicas incorrectly being restarted when joining the writer.
-
Fixed an issue where writes to the database may stall while executing a long running transaction leading to a database restart or a major version upgrade to fail.