Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Duplicate
-
11.4.2
-
None
-
MacOS Sonoma 14.5 (23F79), Homebrew version of MariaDB 11.4.2
Description
Without making any other changes, I just updated MariaDB from 11.3.2 to 11.4.2 and after the restart my database got broken. Initially I thought the problem lies in the update process, but then I managed to reproduce the issue on a clean 11.4.2.
Steps to reproduce :
1. Install a fresh 11.4.2 or just delete the db files and do mariadb-install-db, doesn't matter (At this stage the DB works even after stopping and starting it back).
2. Any kind of small operations like creating/removing users etc are fine (still works after stop and start)
3. Import some big SQL file (a 2GB database dump in my case).
After doing the above and without stopping the database - it continues to work properly, BUT once I stop it and start it back I get this (this is the full log, one successful start, then shutdown and then the failure) :
240610 16:13:02 mysqld_safe Starting mariadbd daemon with databases from /opt/homebrew/var/mysql
|
2024-06-10 16:13:02 0 [Warning] Setting lower_case_table_names=2 because file system for /opt/homebrew/var/mysql/ is case insensitive
|
2024-06-10 16:13:02 0 [Note] Starting MariaDB 11.4.2-MariaDB source revision 3fca5ed772fb75e3e57c507edef2985f8eba5b12 as process 17229
|
2024-06-10 16:13:02 0 [Note] InnoDB: Compressed tables use zlib 1.2.12
|
2024-06-10 16:13:02 0 [Note] InnoDB: Number of transaction pools: 1
|
2024-06-10 16:13:02 0 [Note] InnoDB: Using generic crc32 instructions
|
2024-06-10 16:13:02 0 [Note] InnoDB: Initializing buffer pool, total size = 128.000MiB, chunk size = 2.000MiB
|
2024-06-10 16:13:02 0 [Note] InnoDB: Completed initialization of buffer pool
|
2024-06-10 16:13:02 0 [Note] InnoDB: End of log at LSN=47695
|
2024-06-10 16:13:02 0 [Note] InnoDB: Opened 3 undo tablespaces
|
2024-06-10 16:13:02 0 [Note] InnoDB: 128 rollback segments in 3 undo tablespaces are active.
|
2024-06-10 16:13:02 0 [Note] InnoDB: Setting file './ibtmp1' size to 12.000MiB. Physically writing the file full; Please wait ...
|
2024-06-10 16:13:02 0 [Note] InnoDB: File './ibtmp1' size is now 12.000MiB.
|
2024-06-10 16:13:02 0 [Note] InnoDB: log sequence number 47695; transaction id 14
|
2024-06-10 16:13:02 0 [Note] InnoDB: Loading buffer pool(s) from /opt/Homebrew/var/mysql/ib_buffer_pool
|
2024-06-10 16:13:02 0 [Note] Plugin 'FEEDBACK' is disabled.
|
2024-06-10 16:13:02 0 [Note] Plugin 'wsrep-provider' is disabled.
|
2024-06-10 16:13:02 0 [Note] InnoDB: Buffer pool(s) load completed at 240610 16:13:02
|
2024-06-10 16:13:03 0 [Note] Server socket created on IP: '127.0.0.1'.
|
2024-06-10 16:13:03 0 [Warning] 'user' entry '@joro.home' ignored in --skip-name-resolve mode.
|
2024-06-10 16:13:03 0 [Warning] 'proxies_priv' entry '@% joro@joro.home' ignored in --skip-name-resolve mode.
|
2024-06-10 16:13:03 0 [Note] mariadbd: Event Scheduler: Loaded 0 events
|
2024-06-10 16:13:03 0 [Note] /opt/homebrew/opt/mariadb/bin/mariadbd: ready for connections.
|
Version: '11.4.2-MariaDB' socket: '/tmp/mysql.sock' port: 3306 Homebrew
|
2024-06-10 16:57:35 0 [Note] /opt/homebrew/opt/mariadb/bin/mariadbd (initiated by: unknown): Normal shutdown
|
2024-06-10 16:57:35 0 [Note] InnoDB: FTS optimize thread exiting.
|
2024-06-10 16:57:35 0 [Note] InnoDB: Starting shutdown...
|
2024-06-10 16:57:35 0 [Note] InnoDB: Dumping buffer pool(s) to /opt/Homebrew/var/mysql/ib_buffer_pool
|
2024-06-10 16:57:35 0 [Note] InnoDB: Restricted to 2000 pages due to innodb_buf_pool_dump_pct=25
|
2024-06-10 16:57:35 0 [Note] InnoDB: Buffer pool(s) dump completed at 240610 16:57:35
|
2024-06-10 16:57:36 0 [Note] InnoDB: Removed temporary tablespace data file: "./ibtmp1"
|
2024-06-10 16:57:36 0 [Note] InnoDB: Shutdown completed; log sequence number 10514036011; transaction id 105174
|
2024-06-10 16:57:36 0 [Note] /opt/homebrew/opt/mariadb/bin/mariadbd: Shutdown complete
|
|
240610 16:57:40 mysqld_safe Starting mariadbd daemon with databases from /opt/homebrew/var/mysql
|
2024-06-10 16:57:40 0 [Warning] Setting lower_case_table_names=2 because file system for /opt/homebrew/var/mysql/ is case insensitive
|
2024-06-10 16:57:40 0 [Note] Starting MariaDB 11.4.2-MariaDB source revision 3fca5ed772fb75e3e57c507edef2985f8eba5b12 as process 20792
|
2024-06-10 16:57:40 0 [Note] InnoDB: Compressed tables use zlib 1.2.12
|
2024-06-10 16:57:40 0 [Note] InnoDB: Number of transaction pools: 1
|
2024-06-10 16:57:40 0 [Note] InnoDB: Using generic crc32 instructions
|
2024-06-10 16:57:40 0 [Note] InnoDB: Initializing buffer pool, total size = 128.000MiB, chunk size = 2.000MiB
|
2024-06-10 16:57:40 0 [Note] InnoDB: Completed initialization of buffer pool
|
2024-06-10 16:57:40 0 [ERROR] InnoDB: Missing FILE_CHECKPOINT(10514035995) at 10514035995
|
2024-06-10 16:57:40 0 [ERROR] InnoDB: Log scan aborted at LSN 10514035995
|
2024-06-10 16:57:40 0 [ERROR] InnoDB: Plugin initialization aborted with error Generic error
|
2024-06-10 16:57:40 0 [Note] InnoDB: Starting shutdown...
|
2024-06-10 16:57:40 0 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
|
2024-06-10 16:57:40 0 [Note] Plugin 'FEEDBACK' is disabled.
|
2024-06-10 16:57:40 0 [Note] Plugin 'wsrep-provider' is disabled.
|
2024-06-10 16:57:40 0 [ERROR] Unknown/unsupported storage engine: InnoDB
|
2024-06-10 16:57:40 0 [ERROR] Aborting
|
240610 16:57:40 mysqld_safe mysqld from pid file /opt/homebrew/var/mysql/joro.home.pid ended
|
My config file is :
[client]
|
port=3306
|
|
[mysqld]
|
sql-mode="ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION"
|
skip-name-resolve
|
skip-host-cache
|
character_set_server = utf8mb4
|
collation_server = utf8mb4_unicode_ci
|
|
bind-address=127.0.0.1
|
port=3306
|
I tried this many times with different configurations but same result. My feeling is the problem arises when :
1. The ib_logfile0 get's filled with data
2. The database gets shut down.
Restarting before the ib_logfile0 gets filled in looks to do no damage. Once it gets broken, only innodb_force_recovery=6 allows it to start and I'm able to dump data from it, BUT in some cases only the indexes are damaged, in others - there is real data loss.
Update : reverted to 11.3.2 - problem gone, everything works fine.
Attachments
Issue Links
- relates to
-
MDEV-34422 InnoDB writes corrupted log on macOS and AIX due to uninitialized log_sys.lsn_lock
- Closed