[MDEV-27713] Crash after a conflict of applier thread with stored procedure call by event scheduler Created: 2022-02-01  Updated: 2023-05-12  Resolved: 2022-03-18

Status: Closed
Project: MariaDB Server
Component/s: Galera, Stored routines
Affects Version/s: 10.5.11, 10.4
Fix Version/s: 10.4.25, 10.5.16, 10.6.8, 10.7.4

Type: Bug Priority: Major
Reporter: Valerii Kravchuk Assignee: Jan Lindström (Inactive)
Resolution: Fixed Votes: 0
Labels: None
Environment:

CentOS 7.9


Issue Links:
Relates
relates to MDEV-24143 Galera nodes "randomly" crashing in I... Closed
relates to MDEV-25037 SIGSEGV in MDL_lock::hog_lock_types_b... Closed

 Description   

MariaDB Server 10.5.11 crashes as follows on CentOS 7.9:

Jan 24 15:27:10 host mariadbd: 2022-01-24 15:27:10 0 [Note] WSREP: cluster conflict due to high priority abort for threads:
Jan 24 15:27:10 host mariadbd: 2022-01-24 15:27:10 0 [Note] WSREP: Winning thread:
Jan 24 15:27:10 host mariadbd: THD: 17, mode: high priority, state: exec, conflict: executing, seqno: 2349780871
Jan 24 15:27:10 host mariadbd: SQL: INSERT INTO some_table
Jan 24 15:27:10 host mariadbd: (col1, col2, ...)
Jan 24 15:27:10 host mariadbd: VALUES (val1, val2, ... )
Jan 24 15:27:10 host mariadbd: 2022-01-24 15:27:10 0 [Note] WSREP: Victim thread:
Jan 24 15:27:10 host mariadbd: THD: 2704636, mode: local, state: exec, conflict: executing, seqno: -1
Jan 24 15:27:10 host mariadbd: SQL: UPDATE `some_table
Jan 24 15:27:10 host mariadbd: SET col3 = 1
Jan 24 15:27:10 host mariadbd: WHERE col3 = 0
Jan 24 15:27:10 host mariadbd: AND id >= NAME_CONST('some_id',8682004)
Jan 24 15:27:10 host mariadbd: AND id < NAME_CONST('lsome_other_id',8682004) + NAME_CONST('some_size',1000)
Jan 24 15:27:10 host mariadbd: 2022-01-24 15:27:10 0 [Note] WSREP: context: /home/buildbot/buildbot/padding_for_CPACK_RPM_BUILD_SOURCE_DIRS_PREFIX/mariadb-10.5.11/storage/innobase/handler/ha_innodb.cc:18724
...
Jan 24 15:27:10 host mariadbd: 220124 15:27:10 [ERROR] mysqld got signal 11 ;
Jan 24 15:27:10 host mariadbd: This could be because you hit a bug. It is also possible that this binary
Jan 24 15:27:10 host mariadbd: or one of the libraries it was linked against is corrupt, improperly built,
Jan 24 15:27:10 host mariadbd: or misconfigured. This error can also be caused by malfunctioning hardware.
Jan 24 15:27:10 host mariadbd: To report this bug, see https://mariadb.com/kb/en/reporting-bugs
Jan 24 15:27:10 host mariadbd: We will try our best to scrape up some info that will hopefully help
Jan 24 15:27:10 host mariadbd: diagnose the problem, but since we have already crashed,
Jan 24 15:27:10 host mariadbd: something is definitely wrong and this may fail.
Jan 24 15:27:10 host mariadbd: Server version: 10.5.11-MariaDB-log
Jan 24 15:27:10 host mariadbd: key_buffer_size=134217728
Jan 24 15:27:10 host mariadbd: read_buffer_size=131072
Jan 24 15:27:10 host mariadbd: max_used_connections=6421
Jan 24 15:27:10 host mariadbd: max_threads=100002
Jan 24 15:27:10 host mariadbd: thread_count=1164
Jan 24 15:27:10 host mariadbd: It is possible that mysqld could use up to
Jan 24 15:27:10 host mariadbd: key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 219323278 K bytes of memory
Jan 24 15:27:10 host mariadbd: Hope that's ok; if not, decrease some variables in the equation.
Jan 24 15:27:10 host mariadbd: Thread pointer: 0x7ef8942caac8
Jan 24 15:27:10 host mariadbd: Attempting backtrace. You can use the following information to find out
Jan 24 15:27:10 host mariadbd: where mysqld died. If you see no messages after this, something went
Jan 24 15:27:10 host mariadbd: terribly wrong...
Jan 24 15:27:10 host mariadbd: stack_bottom = 0x7f0844ef6a5f thread_stack 0x49000
Jan 24 15:27:10 host mariadbd: ??:0(my_print_stacktrace)[0x56451470179e]
Jan 24 15:27:10 host mariadbd: ??:0(handle_fatal_signal)[0x564514106457]
...
Jan 24 15:27:11 host mariadbd: sigaction.c:0(__restore_rt)[0x7f099808f630]
Jan 24 15:27:11 host mariadbd: ??:0(MDL_lock::reschedule_waiters())[0x564513ffef9b]
Jan 24 15:27:11 host mariadbd: ??:0(MDL_lock::remove_ticket(LF_PINS*, MDL_lock::Ticket_list MDL_lock::*, MDL_ticket*))[0x564513fff1b4]
Jan 24 15:27:11 host mariadbd: ??:0(MDL_context::release_lock(enum_mdl_duration, MDL_ticket*))[0x564513fffac4]
Jan 24 15:27:11 host mariadbd: ??:0(Item_func_release_lock::val_int())[0x564514173b4b]
Jan 24 15:27:11 host mariadbd: ??:0(Item_string_with_introducer::is_cs_specified() const)[0x564513eb6d65]
Jan 24 15:27:11 host mariadbd: ??:0(mysql_do(THD*, List<Item>&))[0x564514268039]
Jan 24 15:27:11 host mariadbd: ??:0(mysql_execute_command(THD*))[0x564513f03361]
Jan 24 15:27:11 host mariadbd: ??:0(sp_instr_stmt::exec_core(THD*, unsigned int*))[0x564513e5a506]
Jan 24 15:27:11 host mariadbd: ??:0(sp_lex_keeper::reset_lex_and_exec_core(THD*, unsigned int*, bool, sp_instr*))[0x564513e62fe2]
Jan 24 15:27:11 host mariadbd: ??:0(sp_instr_stmt::execute(THD*, unsigned int*))[0x564513e639dc]
Jan 24 15:27:11 host mariadbd: ??:0(sp_head::execute(THD*, bool))[0x564513e5e524]
Jan 24 15:27:11 host mariadbd: ??:0(sp_head::execute_procedure(THD*, List<Item>*))[0x564513e5fc1f]
Jan 24 15:27:11 host mariadbd: ??:0(comp_ne_creator(bool))[0x564513ef6aca]
Jan 24 15:27:11 host mariadbd: ??:0(Sql_cmd_call::execute(THD*))[0x564513efb48a]
Jan 24 15:27:11 host mariadbd: ??:0(mysql_execute_command(THD*))[0x564513f032be]
Jan 24 15:27:11 host mariadbd: ??:0(sp_instr_stmt::exec_core(THD*, unsigned int*))[0x564513e5a506]
Jan 24 15:27:11 host mariadbd: ??:0(sp_lex_keeper::reset_lex_and_exec_core(THD*, unsigned int*, bool, sp_instr*))[0x564513e62fe2]
Jan 24 15:27:11 host mariadbd: ??:0(sp_instr_stmt::execute(THD*, unsigned int*))[0x564513e639dc]
Jan 24 15:27:11 host mariadbd: ??:0(sp_head::execute(THD*, bool))[0x564513e5e524]
Jan 24 15:27:11 host mariadbd: ??:0(sp_head::execute_procedure(THD*, List<Item>*))[0x564513e5fc1f]
Jan 24 15:27:11 host mariadbd: ??:0(Event_job_data::execute(THD*, bool))[0x564513fda28f]
Jan 24 15:27:11 host mariadbd: ??:0(Event_worker_thread::run(THD*, Event_queue_element_for_exec*))[0x564514285f88]
Jan 24 15:27:11 host mariadbd: ??:0(event_worker_thread)[0x56451428603a]
Jan 24 15:27:11 host mariadbd: ??:0(MyCTX_nopad::finish(unsigned char*, unsigned int*))[0x56451435356d]
Jan 24 15:27:12 host mariadbd: pthread_create.c:0(start_thread)[0x7f0998087ea5]
Jan 24 15:27:12 host kernel: traps: mariadbd[92203] general protection ip:7f0997526712 sp:7efb3dd09030 error:0 in libc-2.17.so[7f09974a4000+1c4000]
Jan 24 15:27:14 host systemd: mariadb.service: main process exited, code=killed, status=11/SEGV 

I can not find any MDEV with similar stack trace.


Generated at Thu Feb 08 09:55:00 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.