Details
-
Bug
-
Status: Closed (View Workflow)
-
Critical
-
Resolution: Duplicate
-
11.4.7
-
OS/Distro: Debian 12
Architecture: x86_64
Install method: Plesk-managed MariaDB, stock Debian package
Memory: 64 GB RAM
InnoDB: default configuration
Plugins: None explicitly enabled other than InnoDB
Socket status before crash: /run/mysqld/mysqld.sock missing after hang
Recovery: Restarted successfully, crash recovery completedOS/Distro: Debian 12 Architecture: x86_64 Install method: Plesk-managed MariaDB, stock Debian package Memory: 64 GB RAM InnoDB: default configuration Plugins: None explicitly enabled other than InnoDB Socket status before crash: /run/mysqld/mysqld.sock missing after hang Recovery: Restarted successfully, crash recovery completed
-
Not for Release Notes
Description
(This is my first bug report here, apologies in advance for any mistakes I make)
MariaDB 11.4.7 crashed with a fatal malloc(): unsorted double linked list corrupted error and signal 6. This caused the server to hang during shutdown (status: "Shutdown in progress"), required a manual kill -9, and forced crash recovery on restart.
The system was stable for over 8 hours before the crash occurred. Load was moderate (~30% CPU) and no configuration changes had recently been made.
This appears to be a glibc memory corruption issue potentially triggered by InnoDB or internal memory management routines. There is no indication of query abuse or external corruption.
Attachments
Issue Links
- duplicates
-
-
- Closed
-