[MDEV-11290] Crash at MYSQL_BIN_LOG::trx_group_commit_leader Created: 2016-11-16  Updated: 2020-08-25  Resolved: 2018-10-16

Status: Closed
Project: MariaDB Server
Component/s: Replication
Affects Version/s: 10.1.16
Fix Version/s: 10.1.16

Type: Bug Priority: Major
Reporter: Will Fong Assignee: Andrei Elkin
Resolution: Cannot Reproduce Votes: 1
Labels: None


 Description   

A customer has reported a crash on a Galera node:

161113  1:30:29 [ERROR] mysqld got signal 11 ;
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: 10.1.16-MariaDB
key_buffer_size=16777216
read_buffer_size=262144
max_used_connections=561
max_threads=1002
thread_count=40
It is possible that mysqld could use up to
key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 806424 K  bytes of memory
Hope that's ok; if not, decrease some variables in the equation.
 
Thread pointer: 0x0x7ff4146c4008
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 = 0x7ff20228c208 thread_stack 0x48400
(my_addr_resolve failure: fork)
/<redacted>/bin/mysqld(my_print_stacktrace+0x2e) [0xc099ee]
/<redacted>/bin/mysqld(handle_fatal_signal+0x4ba) [0x7664da]
/lib64/libpthread.so.0() [0x3fe2a0f7e0]
/<redacted>/bin/mysqld(MYSQL_BIN_LOG::trx_group_commit_leader(MYSQL_BIN_LOG::group_commit_entry*)+0x639) [0x843a59]
/<redacted>/bin/mysqld(MYSQL_BIN_LOG::write_transaction_to_binlog_events(MYSQL_BIN_LOG::group_commit_entry*)+0x1c1) [0x844081]
/<redacted>/bin/mysqld(MYSQL_BIN_LOG::write_transaction_to_binlog(THD*, binlog_cache_mngr*, Log_event*, bool, bool, bool)+0x191) [0x8444b1]
/<redacted>/bin/mysqld() [0x844549]
/<redacted>/bin/mysqld(MYSQL_BIN_LOG::log_and_order(THD*, unsigned long long, bool, bool, bool)+0xa7) [0x844847]
/<redacted>/bin/mysqld(ha_commit_trans(THD*, bool)+0x376) [0x76d356]
/<redacted>/bin/mysqld(trans_commit_stmt(THD*)+0x2b) [0x69dd2b]
/<redacted>/bin/mysqld(mysql_execute_command(THD*)+0xea5) [0x5b2445]
/<redacted>/bin/mysqld(mysql_parse(THD*, char*, unsigned int, Parser_state*)+0x2b4) [0x5baeb4]
/<redacted>/bin/mysqld() [0x5baf48]
/<redacted>/bin/mysqld(dispatch_command(enum_server_command, THD*, char*, unsigned int)+0x14e5) [0x5bd025]
/<redacted>/bin/mysqld(do_command(THD*)+0x2a7) [0x5be2f7]
/<redacted>/bin/mysqld(do_handle_one_connection(THD*)+0x183) [0x68d423]
/<redacted>/bin/mysqld(handle_one_connection+0x42) [0x68d642]
/lib64/libpthread.so.0() [0x3fe2a07aa1]
/lib64/libc.so.6(clone+0x6d) [0x3fe22e8aad]
 
Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0x7ff440561020): INSERT INTO <** redacted 65+KB INSERT statement **>
 
Connection ID (thread ID): 3622613
Status: NOT_KILLED
 
Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,deri
ved_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join
_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on,orderby_uses_equalities=off
 
The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains
information that should help you find out what is causing the crash.
161113 01:30:29 mysqld_safe Number of processes running now: 0
161113 01:30:29 mysqld_safe WSREP: not restarting wsrep node automatically
161113 01:30:29 mysqld_safe mysqld from pid file /<redacted>/logs/mariadb.pid ended

This crash has only happened once, during normal operations. There didn't seem to be anything special going on. The node was able to be brought back into the cluster without a problem.



 Comments   
Comment by Daniel Black [ 2017-02-25 ]

Is it possible to use addr2line (binutils) to get line numbers for critical elements of the backtrace?

Comment by Daniel Black [ 2017-02-26 ]

And more importantly, to get the entire context of MYSQL_BIN_LOG::trx_group_commit_leader (as this appears to be a customer compiled code)

gdb `which mysqld` -batch -ex 'disassemble /m MYSQL_BIN_LOG::trx_group_commit_leader'

Or failing that (less pretty as it doesn't include source):

objdump -dCl --start-address 0x0x843420 --stop-address  0x8445b1 `which mysqld`

Comment by Andrei Elkin [ 2018-10-16 ]

The reporter has not provided any method to reproduce and seems to have become
unavailable to provide necessary details for analysis. Considering that and also 2 years elapsed (to infer a number of bug fixes has been done) I am closing this ticket.

Generated at Thu Feb 08 07:48:48 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.