Details
-
Bug
-
Status: Closed (View Workflow)
-
Major
-
Resolution: Incomplete
-
10.6.21
-
None
-
Red Hat Enterprise Linux 8.2 / Kernel : 4.18.0-193.el8.x86_64
Description
I upgraded a MariaDB 10.5.16 node to 10.6.21
On startup, the node crashed with:
250410 18:52:23 mysqld_safe mysqld from pid file /mydata/10/mariadb.pid ended
250410 18:59:16 mysqld_safe Starting mariadbd daemon with databases from /mydata/10/data/
2025-04-10 18:59:16 0 [Note] Starting MariaDB 10.6.21-MariaDB-log source revision 066e8d6aeabc13242193780341e0f845528105de server_uid gMIrqoPrmJ52CK8uHm6MKRUUzmw= as process 61308
2025-04-10 18:59:16 0 [Warning] Could not increase number of max_open_files to more than 65535 (request: 245040)
...
2025-04-10 18:59:17 0 [ERROR] Incorrect definition of table mysql.event: expected column 'definer' at position 3 to have type varchar(, found type char(141).
2025-04-10 18:59:17 0 [ERROR] mariadbd: Event Scheduler: An error occurred when initializing system tables. Disabling the Event Scheduler.
2025-04-10 18:59:17 0 [Note] /maria/10/bin/mariadbd: ready for connections.
Version: '10.6.21-MariaDB-log' socket: '/maria/10/etc/maria.sock' port: 2382 Source distribution
2025-04-10 19:00:04 0 [Warning] InnoDB: Double free of page 163877 in file /mylog/10/undo/undo001
2025-04-10 19:00:21 99 [Note] Event Scheduler: scheduler thread started with id 99
2025-04-10 19:00:23 105 [Warning] 'proxies_priv' entry '@% root@nitndb1d' ignored in --skip-name-resolve mode.
2025-04-10 19:02:08 0 [Warning] InnoDB: Double free of page 163843 in file /mylog/10/undo/undo001
2025-04-10 19:03:39 0 [Warning] InnoDB: Double free of page 163865 in file /mylog/10/undo/undo001
2025-04-10 19:03:49 0 [Warning] InnoDB: Double free of page 163888 in file /mylog/10/undo/undo001
2025-04-10 19:03:50 0 [ERROR] InnoDB: File '/mylog/10/undo/undo002' is corrupted
...
2025-04-10 19:26:29 0x7fc46e2e9700 InnoDB: Assertion failure in file /tmp/server/storage/innobase/trx/trx0purge.cc line 270
InnoDB: Failing assertion: flst_add_first(rseg_header, TRX_RSEG + TRX_RSEG_HISTORY, undo_page, uint16_t(undo_header_offset + TRX_UNDO_HISTORY_NODE), rseg->space->free_limit, mtr) == DB_SUCCESS
InnoDB: We intentionally generate a memory trap.
InnoDB: Submit a detailed bug report to https://jira.mariadb.org/
InnoDB: If you get repeated assertion failures or crashes, even
InnoDB: immediately after the mariadbd startup, there may be
InnoDB: corruption in the InnoDB tablespace. Please refer to
InnoDB: https://mariadb.com/kb/en/library/innodb-recovery-modes/
InnoDB: about forcing recovery.
250410 19:26:29 [ERROR] /maria/10/bin/mariadbd got signal 6 ;
I followed these steps:
- Stop 10.6.21 MariaDB : $MARIA_HOME/bin/mysqladmin shutdown
- Full backup using mariadb-backup from 10.5.16
- Restoring
- Start 10.6.21 MariaDB : $MARIA_HOME/bin/mysqld_safe &
- Upgrade MariaDB 10.5.16 to 10.6.21
At the moment, my upgrate is stuck.
Could you please help me identify the root cause of this issue?
I'm considering moving to version 10.6.20 as a more stable alternative. Does that sound like the right approach?"