Why is mysqld crashing?

Question

linux

InnoDB: Warning: a long semaphore wait:
--Thread 140561570850560 has waited at log0log.cc line 2315 for 524.00 seconds the semaphore:
S-lock on RW-latch at 0x7fd78686e7b8 '&log_sys->checkpoint_lock'
a writer (thread id 140561570850560) has reserved it in mode exclusive
number of readers 0, waiters flag 1, lock_word: 0
Last time read locked in file log0log.cc line 2315
Last time write locked in file log0log.cc line 2077
Holder thread 0 file not yet reserved line 0
InnoDB: Warning: semaphore wait:
--Thread 140561570850560 has waited at log0log.cc line 2315 for 525.00 seconds the semaphore:
S-lock on RW-latch at 0x7fd78686e7b8 '&log_sys->checkpoint_lock'
a writer (thread id 140561570850560) has reserved it in mode exclusive
number of readers 0, waiters flag 1, lock_word: 0
Last time read locked in file log0log.cc line 2315
Last time write locked in file log0log.cc line 2077
Holder thread 0 file not yet reserved line 0
InnoDB: Warning: Writer thread is waiting this semaphore:
--Thread 140561570850560 has waited at log0log.cc line 2315 for 525.00 seconds the semaphore:
S-lock on RW-latch at 0x7fd78686e7b8 '&log_sys->checkpoint_lock'
a writer (thread id 140561570850560) has reserved it in mode exclusive
number of readers 0, waiters flag 1, lock_word: 0
Last time read locked in file log0log.cc line 2315
Last time write locked in file log0log.cc line 2077
Holder thread 0 file not yet reserved line 0
InnoDB: ###### Starts InnoDB Monitor for 30 secs to print diagnostic info:
InnoDB: Pending reads 0, writes 0
InnoDB: ###### Diagnostic info printed to the standard error stream
2021-03-04 4:05:18 140561667290880 [ERROR] InnoDB: The OS said file flush did not succeed
2021-03-04 04:05:18 7fd7103f7700 InnoDB: Operating system error number 1 in a file operation.
InnoDB: Error number 1 means 'Operation not permitted'.
InnoDB: Some operating system error numbers are described at
InnoDB: http://dev.mysql.com/doc/refman/5.6/en/operating-system-error-codes.html
 InnoDB: Operation flush to file /build/mariadb-10.1-53Dg5K/mariadb-10.1-10.1.41/storage/xtradb/os/os0file.cc and at line 2949
2021-03-04 4:05:18 140561667290880 [ERROR] InnoDB: File (unknown): 'flush' returned OS error 201. Cannot continue operation
210304 4:05:18 [ERROR] mysqld got signal 6 ;

Thread pointer: 0x0
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0x0 thread_stack 0x30000
/usr/sbin/mysqld(my_print_stacktrace+0x2e)[0x564a0209ebee]
/usr/sbin/mysqld(handle_fatal_signal+0x3bd)[0x564a01bdf45d]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x110e0)[0x7fd7898960e0]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcf)[0x7fd788403fff]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x16a)[0x7fd78840542a]
/usr/sbin/mysqld(+0x8a006c)[0x564a01ecc06c]
/usr/sbin/mysqld(+0x8a3fb0)[0x564a01ecffb0]
/usr/sbin/mysqld(+0x9e62f8)[0x564a020122f8]
/usr/sbin/mysqld(+0x9e6824)[0x564a02012824]
/usr/sbin/mysqld(+0x993705)[0x564a01fbf705]
/usr/sbin/mysqld(+0x99889f)[0x564a01fc489f]
/usr/sbin/mysqld(+0x98dd28)[0x564a01fb9d28]
/usr/sbin/mysqld(+0x9e0f8b)[0x564a0200cf8b]
/usr/sbin/mysqld(+0x91de00)[0x564a01f49e00]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x74a4)[0x7fd78988c4a4]
/lib/x86_64-linux-gnu/libc.so.6(clone+0x3f)[0x7fd7884b9d0f]
The manual page at dev.mysql.com/doc/mysql/en/crashing.html contains
information that should help you find out what is causing the crash.
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Using mutexes to ref count buffer pool pages
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: The InnoDB memory heap is disabled
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: GCC builtin __atomic_thread_fence() is used for memory barrier
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Compressed tables use zlib 1.2.8
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Using Linux native AIO
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Using SSE crc32 instructions
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Initializing buffer pool, size = 1.0G
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Completed initialization of buffer pool
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Highest supported file format is Barracuda.
2021-03-04 4:05:26 139836899855744 [Note] InnoDB: Starting crash recovery from checkpoint LSN=290193987981
2021-03-04 4:05:29 139836899855744 [Note] InnoDB: Restoring possible half-written data pages from the doublewrite buffer...
2021-03-04 4:05:29 139836899855744 [Note] InnoDB: Starting final batch to recover 202 pages from redo log
2021-03-04 4:05:31 139836899855744 [Note] InnoDB: 128 rollback segment(s) are active.
2021-03-04 4:05:31 139836899855744 [Note] InnoDB: Waiting for purge to start
2021-03-04 4:05:31 139836899855744 [Note] InnoDB: Percona XtraDB (percona.com) 5.6.44-86.0 started; log sequence number 290194011212
2021-03-04 4:05:32 139836899855744 [ERROR] InnoDB: The OS said file flush did not succeed
2021-03-04 04:05:32 7f2e50be0d80 InnoDB: Operating system error number 1 in a file operation.
InnoDB: Error number 1 means 'Operation not permitted'.
InnoDB: Some operating system error numbers are described at
InnoDB: dev.mysql.com/doc/refman/5.6/en/operating-system-error-codes.html
InnoDB: Operation flush to file /build/mariadb-10.1-53Dg5K/mariadb-10.1-10.1.41/storage/xtradb/os/os0file.cc and at line 2949
2021-03-04 4:05:32 139836899855744 [ERROR] InnoDB: File (unknown): 'flush' returned OS error 201. Cannot continue operation
210304 4:05:32 [ERROR] mysqld got signal 6 ;

Server version: 10.1.41-MariaDB-0+deb9u1
key_buffer_size=268435456
read_buffer_size=131072
max_used_connections=0
max_threads=502
thread_count=0
It is possible that mysqld could use up to 
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1364903 K bytes of memory
Hope that's ok; if not, decrease some variables in the equation.

Thread pointer: 0x0
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 0x0 thread_stack 0x30000
*** buffer overflow detected ***: /usr/sbin/mysqld terminated
===== Backtrace: =======
/lib/x86_64-linux-gnu/libc.so.6(+0x70bfb)[0x7f2e4f3f9bfb]
/lib/x86_64-linux-gnu/libc.so.6(__fortify_fail+0x37)[0x7f2e4f482437]
/lib/x86_64-linux-gnu/libc.so.6(+0xf7570)[0x7f2e4f480570]
/lib/x86_64-linux-gnu/libc.so.6(+0xf93aa)[0x7f2e4f4823aa]
/usr/sbin/mysqld(my_addr_resolve+0xd8)[0x562c88780968]
/usr/sbin/mysqld(my_print_stacktrace+0x1bb)[0x562c88769d7b]
/usr/sbin/mysqld(handle_fatal_signal+0x3bd)[0x562c882aa45d]
/lib/x86_64-linux-gnu/libpthread.so.0(+0x110e0)[0x7f2e5084e0e0]
/lib/x86_64-linux-gnu/libc.so.6(gsignal+0xcf)[0x7f2e4f3bbfff]
/lib/x86_64-linux-gnu/libc.so.6(abort+0x16a)[0x7f2e4f3bd42a]
/usr/sbin/mysqld(+0x8a006c)[0x562c8859706c]
/usr/sbin/mysqld(+0x8a3fb0)[0x562c8859afb0]
/usr/sbin/mysqld(+0x9e62f8)[0x562c886dd2f8]
/usr/sbin/mysqld(+0x9e63dc)[0x562c886dd3dc]
/usr/sbin/mysqld(+0x993d8c)[0x562c8868ad8c]
/usr/sbin/mysqld(+0x99d028)[0x562c88694028]
/usr/sbin/mysqld(+0x8787e4)[0x562c8856f7e4]
/usr/sbin/mysqld(+0x8804c8)[0x562c885774c8]
/usr/sbin/mysqld(+0x922a4a)[0x562c88619a4a]
/usr/sbin/mysqld(+0x835de1)[0x562c8852cde1]
/usr/sbin/mysqld(_Z24ha_initialize_handlertonP13st_plugin_int+0x62)[0x562c882ac652]
/usr/sbin/mysqld(+0x42d115)[0x562c88124115]
/usr/sbin/mysqld(_Z11plugin_initPiPPci+0x7da)[0x562c8812577a]
/usr/sbin/mysqld(+0x38430c)[0x562c8807b30c]
/usr/sbin/mysqld(_Z11mysqld_mainiPPc+0x1bd3)[0x562c8807ee33]
/lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf1)[0x7f2e4f3a92e1]
/usr/sbin/mysqld(_start+0x2a)[0x562c88072fca]
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Using mutexes to ref count buffer pool pages
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: The InnoDB memory heap is disabled
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: GCC builtin __atomic_thread_fence() is used for memory barrier
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Compressed tables use zlib 1.2.8
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Using Linux native AIO
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Using SSE crc32 instructions
2021-03-04 4:05:37 140706713234816 [Note] InnoDB: Initializing buffer pool, size = 1.0G
2021-03-04 4:05:38 140706713234816 [Note] InnoDB: Completed initialization of buffer pool
2021-03-04 4:05:38 140706713234816 [Note] InnoDB: Highest supported file format is Barracuda.
2021-03-04 4:05:38 140706713234816 [Note] InnoDB: Starting crash recovery from checkpoint LSN=290193987981
2021-03-04 4:05:38 140706713234816 [Note] InnoDB: Restoring possible half-written data pages from the doublewrite buffer..
2021-03-04 4:05:38 140706713234816 [Note] InnoDB: Starting final batch to recover 202 pages from redo log
2021-03-04 4:05:39 140706713234816 [Note] InnoDB: 128 rollback segment(s) are active.
2021-03-04 4:05:39 140706713234816 [Note] InnoDB: Waiting for purge to start
2021-03-04 4:05:39 140706713234816 [Note] InnoDB: Percona XtraDB (http://www.percona.com) 5.6.44-86.0 started; log sequence number 290194011222
2021-03-04 4:05:42 140706713234816 [ERROR] InnoDB: The OS said file flush did not succeed
2021-03-04 04:05:42 7ff8d5a8ed80 InnoDB: Operating system error number 1 in a file operation.
InnoDB: Error number 1 means 'Operation not permitted'.
InnoDB: Some operating system error numbers are described at
InnoDB: dev.mysql.com/doc/refman/5.6/en/operating-system-error-codes.html
InnoDB: Operation flush to file /build/mariadb-10.1-53Dg5K/mariadb-10.1-10.1.41/storage/xtradb/os/os0file.cc and at line 2949
2021-03-04 4:05:42 140706713234816 [ERROR] InnoDB: File (unknown): 'flush' returned OS error 201. Cannot continue operation
210304 4:05:42 [ERROR] mysqld got signal 6;

Answer 1

I would
1) check that the server has no problems with hardware, there is space on the disk, etc.
2) updated MariaDB to latest stable 10.5

Answer 2

I won't name the hoster...
Vitaly Karasik was right.
Further correspondence with support:

Hello! Mysqld crashed around 4 am.
Server IP: XXX.XXX.XXX.XXX
My mysql log at the link: http://*.ru/error.log_8.1.gz
Everyone recommends checking the hardware.
Similar issue on MySQL forum: https: //forums.mysql.com/read.php
?22,620521,620521...
https://bugs.mysql.com/bug.php?id=74824
Host response:

Hello!
Yes, there were disk problems on the root server. At the moment everything should be fine, but we recommend that you reboot the server just in case so that it checks the file system.

And there was no notice, not a damn thing. Such an attitude

Answer 3

https://community.apachefriends.org/f/viewtopic.ph...

Open MySQL config file by “Config” -> “my.ini”.
Insert
innodb_flush_method=normal
between
innodb_lock_wait_timeout = 50
and
## UTF 8 Settings
Start MySQL again in xampp.

Update

*** buffer overflow detected ***: /usr/sbin/mysqld terminated

Buffer overflow is a bug in the DBMS itself. For me, you need to either update to the latest fix, if any, or, as Vitaly Karasik advises , to a later version. This may be caused by the environment, settings, etc.
Well, it is better to ask very specific questions directly from the MariaDB developers in English.