[MDEV-26445] innodb_undo_log_truncate is unnecessarily slow - Jira

Details

Type: Bug
Status: Closed (View Workflow)
Priority: Critical
Resolution: Fixed
Affects Version/s: 10.5.11, 10.6
Fix Version/s: 10.5.13, 10.6.5, 10.7.1
Component/s: Storage Engine - InnoDB
Labels:
- performance
Environment:
debian 11

Description

Even after the version 10.5.9 which should have fixed it, it still crashes for me

--Thread 139751148128000 has waited at btr0pcur.cc line 510 for 435.00 seconds the semaphore:

S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221

a writer (thread id 139751159285504) has reserved it in mode  exclusive

number of readers 0, waiters flag 1, lock_word: 0

Last time write locked in file btr0pcur.cc line 257

2021-08-20  7:33:31 0 [Note] InnoDB: A semaphore wait:

--Thread 139751139219200 has waited at btr0pcur.cc line 510 for 254.00 seconds the semaphore:

S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221

a writer (thread id 139751159285504) has reserved it in mode  exclusive

number of readers 0, waiters flag 1, lock_word: 0

Last time write locked in file btr0pcur.cc line 257

2021-08-20  7:33:31 0 [Note] InnoDB: A semaphore wait:

--Thread 139751134918400 has waited at btr0pcur.cc line 510 for 511.00 seconds the semaphore:

S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221

a writer (thread id 139751159285504) has reserved it in mode  exclusive

number of readers 0, waiters flag 1, lock_word: 0

Last time write locked in file btr0pcur.cc line 257

InnoDB: Pending reads 0, writes 0

2021-08-20  7:33:31 0 [ERROR] [FATAL] InnoDB: Semaphore wait has lasted > 600 seconds. We intentionally crash the server because it appears to be hung.

210820  7:33:31 [ERROR] mysqld got signal 6 ;

This could be because you hit a bug. It is also possible that this binary

or one of the libraries it was linked against is corrupt, improperly built,

or misconfigured. This error can also be caused by malfunctioning hardware.

To report this bug, see https://mariadb.com/kb/en/reporting-bugs

We will try our best to scrape up some info that will hopefully help

diagnose the problem, but since we have already crashed,

something is definitely wrong and this may fail.

Server version: 10.5.11-MariaDB-1

key_buffer_size=134217728

read_buffer_size=131072

max_used_connections=346

max_threads=402

thread_count=330

It is possible that mysqld could use up to

key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1016043 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 0x49000

Can't start addr2line

/usr/sbin/mariadbd(my_print_stacktrace+0x2e)[0x555b4482793e]

/usr/sbin/mariadbd(handle_fatal_signal+0x485)[0x555b44333145]

/lib/x86_64-linux-gnu/libpthread.so.0(+0x14140)[0x7f40b1317140]

/lib/x86_64-linux-gnu/libc.so.6(gsignal+0x141)[0x7f40b0e60ce1]

/lib/x86_64-linux-gnu/libc.so.6(abort+0x123)[0x7f40b0e4a537]

/usr/sbin/mariadbd(+0x62fefd)[0x555b44017efd]

/usr/sbin/mariadbd(+0x626b02)[0x555b4400eb02]

/usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic13timer_generic7executeEPv+0x3b)[0x555b447c91fb]

/usr/sbin/mariadbd(_ZN5tpool4task7executeEv+0x32)[0x555b447ca592]

/usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic11worker_mainEPNS_11worker_dataE+0x4f)[0x555b447c8fbf]

/lib/x86_64-linux-gnu/libstdc++.so.6(+0xceed0)[0x7f40b11feed0]

/lib/x86_64-linux-gnu/libpthread.so.0(+0x8ea7)[0x7f40b130bea7]

/lib/x86_64-linux-gnu/libc.so.6(clone+0x3f)[0x7f40b0f22def]

The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains

information that should help you find out what is causing the crash.

Writing a core file...

Working directory at /home/mysql

Resource Limits:

Limit                     Soft Limit           Hard Limit           Units

Max cpu time              unlimited            unlimited            seconds

Max file size             unlimited            unlimited            bytes

Max data size             unlimited            unlimited            bytes

Max stack size            8388608              unlimited            bytes

Max core file size        0                    unlimited            bytes

Max resident set          unlimited            unlimited            bytes

Max processes             1030947              1030947              processes

Max open files            60000                60000                files

Max locked memory         65536                65536                bytes

Max address space         unlimited            unlimited            bytes

Max file locks            unlimited            unlimited            locks

Max pending signals       1030947              1030947              signals

Max msgqueue size         819200               819200               bytes

Max nice priority         0                    0

Max realtime priority     0                    0

Max realtime timeout      unlimited            unlimited            us

Core pattern: core

Then after this crash, it has corrupted undo file and mariadb can't start :

2021-08-20 10:07:23 0 [ERROR] [FATAL] InnoDB: Trying to read 16384 bytes at 884310016 outside the bounds of the file: .//undo002

210820 10:07:23 [ERROR] mysqld got signal 6 ;

This could be because you hit a bug. It is also possible that this binary

or one of the libraries it was linked against is corrupt, improperly built,

or misconfigured. This error can also be caused by malfunctioning hardware.

To report this bug, see https://mariadb.com/kb/en/reporting-bugs

We will try our best to scrape up some info that will hopefully help

diagnose the problem, but since we have already crashed,

something is definitely wrong and this may fail.

Server version: 10.5.11-MariaDB-1

key_buffer_size=134217728

read_buffer_size=131072

max_used_connections=0

max_threads=402

thread_count=0

It is possible that mysqld could use up to

key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 1016043 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 0x49000

??:0(my_print_stacktrace)[0x55a09b42993e]

??:0(handle_fatal_signal)[0x55a09af35145]

sigaction.c:0(__restore_rt)[0x7f9b7a0e0140]

linux/raise.c:51(__GI_raise)[0x7f9b79c29ce1]

stdlib/abort.c:81(__GI_abort)[0x7f9b79c13537]

??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac19efd]

??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d0f2]

??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d137]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34b247]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34c494]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b33adb0]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b302398]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2f873a]

??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2fee6e]

??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac13662]

??:0(wsrep_notify_status(wsrep::server_state::state, wsrep::view const*))[0x55a09b213139]

??:0(ha_initialize_handlerton(st_plugin_int*))[0x55a09af380de]

??:0(sys_var_pluginvar::sys_var_pluginvar(sys_var_chain*, char const*, st_plugin_int*, st_mysql_sys_var*))[0x55a09ad4db5a]

??:0(plugin_init(int*, char**, int))[0x55a09ad4ec47]

??:0(unireg_abort)[0x55a09ac7a457]

??:0(mysqld_main(int, char**))[0x55a09ac800e0]

csu/libc-start.c:308(__libc_start_main)[0x7f9b79c14d0a]

??:0(_start)[0x55a09ac7446a]

The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains

information that should help you find out what is causing the crash.

Writing a core file...

Working directory at /home/mysql

Resource Limits:

Limit                     Soft Limit           Hard Limit           Units

Max cpu time              unlimited            unlimited            seconds

Max file size             unlimited            unlimited            bytes

Max data size             unlimited            unlimited            bytes

Max stack size            8388608              unlimited            bytes

Max core file size        0                    unlimited            bytes

Max resident set          unlimited            unlimited            bytes

Max processes             1030947              1030947              processes

Max open files            60000                60000                files

Max locked memory         65536                65536                bytes

Max address space         unlimited            unlimited            bytes

Max file locks            unlimited            unlimited            locks

Max pending signals       1030947              1030947              signals

Max msgqueue size         819200               819200               bytes

Max nice priority         0                    0

Max realtime priority     0                    0

Max realtime timeout      unlimited            unlimited            us

Core pattern: core

Attachments

- Sort By Name
- Sort By Date
- Ascending
- Descending
- Thumbnails
- List
- Download All

crash.log
996 kB
2021-08-25 08:11
crash2.log
738 kB
2021-08-31 15:24
crash3.log
1.30 MB
2021-09-04 19:03
hopeenoughdata.txt
40 kB
2021-09-15 10:58
log.txt
142 kB
2021-08-25 08:16
mariadb_bt_all_threads_workload_slowed_down.txt
69 kB
2021-09-16 08:31
mariadb_bt_all_threads.txt
19 kB
2021-09-15 19:57
semaphore without crash.txt
713 kB
2021-09-06 08:37
showenginecensored.txt
28 kB
2021-09-09 13:02
show status.txt
62 kB
2021-09-13 16:21
stackfreeze
209 kB
2021-09-09 11:42
stackfreeze2
442 kB
2021-09-09 12:03
stackfreeze3
442 kB
2021-09-09 12:11
stillstuckmariadb.txt
22 kB
2021-09-13 14:37
stillstuckmariadb3
17 kB
2021-09-13 16:21

Issue Links

causes

MDEV-32757 innodb_undo_log_truncate=ON is not crash safe

Closed

includes

MDEV-21952 ibdata1 file size growing in MariaDB

Closed

is blocked by

MDEV-26450 Corruption due to innodb_undo_log_truncate

Closed

relates to

MDEV-24911 Missing warning before [ERROR] [FATAL] InnoDB: innodb_fatal_semaphore_wait_threshold was exceeded for dict_sys.mutex

Open

MDEV-26662 buf_read_page_background() is blocking on LRU eviction

Closed

MDEV-27530 InnoDB - Performance issues after upgrade 10.4.22 to 10.5.13

Closed

(1 relates to)

Activity

Ascending order - Click to sort in descending order

martina342 created issue - 2021-08-20 08:24

martina342 made changes - 2021-08-20 08:26

Field	Original Value	New Value
Description	Even after the version 10.5.9 which should have fixed it, it still crashes for me --Thread 139751148128000 has waited at btr0pcur.cc line 510 for 435.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 2021-08-20 7:33:31 0 [Note] InnoDB: A semaphore wait: --Thread 139751139219200 has waited at btr0pcur.cc line 510 for 254.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 2021-08-20 7:33:31 0 [Note] InnoDB: A semaphore wait: --Thread 139751134918400 has waited at btr0pcur.cc line 510 for 511.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 InnoDB: Pending reads 0, writes 0 2021-08-20 7:33:31 0 [ERROR] [FATAL] InnoDB: Semaphore wait has lasted > 600 seconds. We intentionally crash the server because it appears to be hung. 210820 7:33:31 [ERROR] mysqld got signal 6 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.11-MariaDB-1 key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=346 max_threads=402 thread_count=330 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)max_threads = 1016043 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 0x49000 Can't start addr2line /usr/sbin/mariadbd(my_print_stacktrace+0x2e)[0x555b4482793e] /usr/sbin/mariadbd(handle_fatal_signal+0x485)[0x555b44333145] /lib/x86_64-linux-gnu/libpthread.so.0(+0x14140)[0x7f40b1317140] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x141)[0x7f40b0e60ce1] /lib/x86_64-linux-gnu/libc.so.6(abort+0x123)[0x7f40b0e4a537] /usr/sbin/mariadbd(+0x62fefd)[0x555b44017efd] /usr/sbin/mariadbd(+0x626b02)[0x555b4400eb02] /usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic13timer_generic7executeEPv+0x3b)[0x555b447c91fb] /usr/sbin/mariadbd(_ZN5tpool4task7executeEv+0x32)[0x555b447ca592] /usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic11worker_mainEPNS_11worker_dataE+0x4f)[0x555b447c8fbf] /lib/x86_64-linux-gnu/libstdc++.so.6(+0xceed0)[0x7f40b11feed0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x8ea7)[0x7f40b130bea7] /lib/x86_64-linux-gnu/libc.so.6(clone+0x3f)[0x7f40b0f22def] The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/mysql Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 1030947 1030947 processes Max open files 60000 60000 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 1030947 1030947 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core Then after this crash, it has corrupted undo file and mariadb can't start : 2021-08-20 10:07:23 0 [ERROR] [FATAL] InnoDB: Trying to read 16384 bytes at 884310016 outside the bounds of the file: .//undo002 210820 10:07:23 [ERROR] mysqld got signal 6 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.11-MariaDB-1 key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=0 max_threads=402 thread_count=0 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)max_threads = 1016043 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 0x49000 ??:0(my_print_stacktrace)[0x55a09b42993e] ??:0(handle_fatal_signal)[0x55a09af35145] sigaction.c:0(__restore_rt)[0x7f9b7a0e0140] linux/raise.c:51(__GI_raise)[0x7f9b79c29ce1] stdlib/abort.c:81(__GI_abort)[0x7f9b79c13537] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac19efd] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d0f2] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d137] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34b247] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34c494] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b33adb0] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b302398] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2f873a] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2fee6e] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac13662] ??:0(wsrep_notify_status(wsrep::server_state::state, wsrep::view const))[0x55a09b213139] ??:0(ha_initialize_handlerton(st_plugin_int))[0x55a09af380de] ??:0(sys_var_pluginvar::sys_var_pluginvar(sys_var_chain, char const, st_plugin_int, st_mysql_sys_var))[0x55a09ad4db5a] ??:0(plugin_init(int, char, int))[0x55a09ad4ec47] ??:0(unireg_abort)[0x55a09ac7a457] ??:0(mysqld_main(int, char*))[0x55a09ac800e0] csu/libc-start.c:308(__libc_start_main)[0x7f9b79c14d0a] ??:0(_start)[0x55a09ac7446a] The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/mysql Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 1030947 1030947 processes Max open files 60000 60000 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 1030947 1030947 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core	Even after the version 10.5.9 which should have fixed it, it still crashes for me {noformat} --Thread 139751148128000 has waited at btr0pcur.cc line 510 for 435.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 2021-08-20 7:33:31 0 [Note] InnoDB: A semaphore wait: --Thread 139751139219200 has waited at btr0pcur.cc line 510 for 254.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 2021-08-20 7:33:31 0 [Note] InnoDB: A semaphore wait: --Thread 139751134918400 has waited at btr0pcur.cc line 510 for 511.00 seconds the semaphore: S-lock on RW-latch at 0x7f3e1c0ca778 created in file buf0buf.cc line 1221 a writer (thread id 139751159285504) has reserved it in mode exclusive number of readers 0, waiters flag 1, lock_word: 0 Last time write locked in file btr0pcur.cc line 257 InnoDB: Pending reads 0, writes 0 2021-08-20 7:33:31 0 [ERROR] [FATAL] InnoDB: Semaphore wait has lasted > 600 seconds. We intentionally crash the server because it appears to be hung. 210820 7:33:31 [ERROR] mysqld got signal 6 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.11-MariaDB-1 key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=346 max_threads=402 thread_count=330 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)max_threads = 1016043 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 0x49000 Can't start addr2line /usr/sbin/mariadbd(my_print_stacktrace+0x2e)[0x555b4482793e] /usr/sbin/mariadbd(handle_fatal_signal+0x485)[0x555b44333145] /lib/x86_64-linux-gnu/libpthread.so.0(+0x14140)[0x7f40b1317140] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x141)[0x7f40b0e60ce1] /lib/x86_64-linux-gnu/libc.so.6(abort+0x123)[0x7f40b0e4a537] /usr/sbin/mariadbd(+0x62fefd)[0x555b44017efd] /usr/sbin/mariadbd(+0x626b02)[0x555b4400eb02] /usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic13timer_generic7executeEPv+0x3b)[0x555b447c91fb] /usr/sbin/mariadbd(_ZN5tpool4task7executeEv+0x32)[0x555b447ca592] /usr/sbin/mariadbd(_ZN5tpool19thread_pool_generic11worker_mainEPNS_11worker_dataE+0x4f)[0x555b447c8fbf] /lib/x86_64-linux-gnu/libstdc++.so.6(+0xceed0)[0x7f40b11feed0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x8ea7)[0x7f40b130bea7] /lib/x86_64-linux-gnu/libc.so.6(clone+0x3f)[0x7f40b0f22def] The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/mysql Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 1030947 1030947 processes Max open files 60000 60000 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 1030947 1030947 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core {noformat} Then after this crash, it has corrupted undo file and mariadb can't start : {noformat} 2021-08-20 10:07:23 0 [ERROR] [FATAL] InnoDB: Trying to read 16384 bytes at 884310016 outside the bounds of the file: .//undo002 210820 10:07:23 [ERROR] mysqld got signal 6 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see https://mariadb.com/kb/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.5.11-MariaDB-1 key_buffer_size=134217728 read_buffer_size=131072 max_used_connections=0 max_threads=402 thread_count=0 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)max_threads = 1016043 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 0x49000 ??:0(my_print_stacktrace)[0x55a09b42993e] ??:0(handle_fatal_signal)[0x55a09af35145] sigaction.c:0(__restore_rt)[0x7f9b7a0e0140] linux/raise.c:51(__GI_raise)[0x7f9b79c29ce1] stdlib/abort.c:81(__GI_abort)[0x7f9b79c13537] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac19efd] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d0f2] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac2d137] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34b247] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b34c494] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b33adb0] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b302398] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2f873a] ??:0(std::unique_lock<std::mutex>::unlock())[0x55a09b2fee6e] ??:0(Wsrep_server_service::log_dummy_write_set(wsrep::client_state&, wsrep::ws_meta const&))[0x55a09ac13662] ??:0(wsrep_notify_status(wsrep::server_state::state, wsrep::view const))[0x55a09b213139] ??:0(ha_initialize_handlerton(st_plugin_int))[0x55a09af380de] ??:0(sys_var_pluginvar::sys_var_pluginvar(sys_var_chain, char const, st_plugin_int, st_mysql_sys_var))[0x55a09ad4db5a] ??:0(plugin_init(int, char, int))[0x55a09ad4ec47] ??:0(unireg_abort)[0x55a09ac7a457] ??:0(mysqld_main(int, char*))[0x55a09ac800e0] csu/libc-start.c:308(__libc_start_main)[0x7f9b79c14d0a] ??:0(_start)[0x55a09ac7446a] The manual page at https://mariadb.com/kb/en/how-to-produce-a-full-stack-trace-for-mysqld/ contains information that should help you find out what is causing the crash. Writing a core file... Working directory at /home/mysql Resource Limits: Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size unlimited unlimited bytes Max data size unlimited unlimited bytes Max stack size 8388608 unlimited bytes Max core file size 0 unlimited bytes Max resident set unlimited unlimited bytes Max processes 1030947 1030947 processes Max open files 60000 60000 files Max locked memory 65536 65536 bytes Max address space unlimited unlimited bytes Max file locks unlimited unlimited locks Max pending signals 1030947 1030947 signals Max msgqueue size 819200 819200 bytes Max nice priority 0 0 Max realtime priority 0 0 Max realtime timeout unlimited unlimited us Core pattern: core {noformat}

Marko Mäkelä made changes - 2021-08-20 08:40

Component/s		Storage Engine - InnoDB [ 10129 ]
Labels		need_feedback

martina342 made changes - 2021-08-24 09:03

Labels

need_feedback

martina342 made changes - 2021-08-25 08:11

Attachment

crash.log [ 58683 ]

martina342 made changes - 2021-08-25 08:16

Attachment

log.txt [ 58684 ]

martina342 made changes - 2021-08-31 15:24

Attachment

crash2.log [ 58802 ]

Alice Sherepa made changes - 2021-08-31 16:28

Link

This issue relates to MDEV-24911 [ MDEV-24911 ]

martina342 made changes - 2021-09-04 19:03

Attachment

crash3.log [ 58849 ]

martina342 made changes - 2021-09-06 08:37

Attachment

semaphore without crash.txt [ 58852 ]

martina342 made changes - 2021-09-09 11:42

Attachment

stackfreeze [ 58896 ]

martina342 made changes - 2021-09-09 12:03

Attachment

stackfreeze2 [ 58898 ]

martina342 made changes - 2021-09-09 12:11

Attachment

stackfreeze3 [ 58899 ]

martina342 made changes - 2021-09-09 13:02

Attachment

showenginecensored.txt [ 58901 ]

martina342 made changes - 2021-09-13 14:37

Attachment

stillstuckmariadb.txt [ 58945 ]

martina342 made changes - 2021-09-13 16:21

Attachment

show status.txt [ 58948 ]

martina342 made changes - 2021-09-13 16:21

Attachment

stillstuckmariadb3 [ 58949 ]

martina342 made changes - 2021-09-15 10:58

Attachment

hopeenoughdata.txt [ 58996 ]

martina342 made changes - 2021-09-15 19:49

Attachment

mariadb_bt_all_threads.txt [ 59005 ]

martina342 made changes - 2021-09-15 19:50

Attachment

mariadb_bt_all_threads.txt [ 59005 ]

martina342 made changes - 2021-09-15 19:57

Attachment

mariadb_bt_all_threads.txt [ 59006 ]

martina342 made changes - 2021-09-16 08:31

Attachment

mariadb_bt_all_threads_workload_slowed_down.txt [ 59013 ]

Marko Mäkelä made changes - 2021-09-16 08:58

Link

This issue is blocked by ~~MDEV-26450~~ [ ~~MDEV-26450~~ ]

Marko Mäkelä made changes - 2021-09-16 13:00

Assignee

Marko Mäkelä [ marko ]

Marko Mäkelä made changes - 2021-09-21 15:19

Link

This issue blocks MENT-1304 [ MENT-1304 ]

Marko Mäkelä made changes - 2021-09-21 15:19

Fix Version/s		10.5 [ 23123 ]
Fix Version/s		10.6 [ 24028 ]
Affects Version/s		10.6 [ 24028 ]
Labels		performance
Summary	[ERROR] [FATAL] InnoDB: Semaphore wait has lasted > 600 seconds. We intentionally crash the server because it appears to be hung.	innodb_undo_log_truncate is unnecessarily slow

Marko Mäkelä made changes - 2021-09-21 15:19

Status

Open [ 1 ]

In Progress [ 3 ]

Marko Mäkelä made changes - 2021-09-21 15:22

Link

This issue relates to ~~MDEV-21952~~ [ ~~MDEV-21952~~ ]

Marko Mäkelä made changes - 2021-09-22 06:50

Link

This issue relates to ~~MDEV-26662~~ [ ~~MDEV-26662~~ ]

Marko Mäkelä made changes - 2021-09-24 15:02

issue.field.resolutiondate

2021-09-24 15:02:51.0

2021-09-24 15:02:51.532

Marko Mäkelä made changes - 2021-09-24 15:02

Fix Version/s		10.5.13 [ 26026 ]
Fix Version/s		10.6.5 [ 26034 ]
Fix Version/s		10.7.1 [ 26120 ]
Fix Version/s	10.5 [ 23123 ]
Fix Version/s	10.6 [ 24028 ]
Resolution		Fixed [ 1 ]
Status	In Progress [ 3 ]	Closed [ 6 ]

Sergei Golubchik made changes - 2021-10-25 10:02

Link

This issue includes ~~MDEV-21952~~ [ ~~MDEV-21952~~ ]

Sergei Golubchik made changes - 2021-10-25 10:02

Link

This issue relates to ~~MDEV-21952~~ [ ~~MDEV-21952~~ ]

Sergei Golubchik made changes - 2021-12-06 21:53

Workflow

MariaDB v3 [ 124481 ]

MariaDB v4 [ 159625 ]

Alice Sherepa made changes - 2022-01-18 12:38

Link

This issue relates to ~~MDEV-27530~~ [ ~~MDEV-27530~~ ]

Marko Mäkelä made changes - 2023-11-10 07:49

Link

This issue causes ~~MDEV-32757~~ [ ~~MDEV-32757~~ ]

Jira Automation (IT) made changes - 2024-07-04 02:22

Zendesk Related Tickets

131727

People

Assignee:: Marko Mäkelä

Reporter:: martina342

Votes:: 0 Vote for this issue

Watchers:: 7 Start watching this issue

Dates

Created:: 2021-08-20 08:24

Updated:: 2024-07-07 21:16

Resolved:: 2021-09-24 15:02

Git Integration

Error rendering 'com.xiplink.jira.git.jira_git_plugin:git-issue-webpanel'. Please contact your Jira administrators.

MariaDB Server

Details

Description

Attachments

Attachments

Issue Links

Activity

People

Dates

Git Integration