Bug #67115 MySQL 5.6.7-rc crashed under tpcc-mysql workload
Submitted: 6 Oct 2012 18:49 Modified: 29 Dec 2012 16:33
Reporter: Vadim TKACHENKO Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Server: InnoDB storage engine Severity:S1 (Critical)
Version:5.6.7 OS:Any
Assigned to: CPU Architecture:Any

[6 Oct 2012 18:49] Vadim TKACHENKO
Description:
Under my tpcc-mysql benchmark I've got crash of MySQL 5.6.7

Version: '5.6.7-rc'  socket: '/tmp/mysql.sock'  port: 3306  MySQL Community Server (GPL)                                            
121006  0:10:24  InnoDB: Assertion failure in thread 139567080404736 in file buf0dblwr.cc line 711                                  
InnoDB: Failing assertion: buf_page_in_file(&block->page)                                                                           
InnoDB: We intentionally generate a memory trap.                                                                                    
InnoDB: Submit a detailed bug report to http://bugs.mysql.com.                                                                      
InnoDB: If you get repeated assertion failures or crashes, even                                                                     
InnoDB: immediately after the mysqld startup, there may be                                                                          
InnoDB: corruption in the InnoDB tablespace. Please refer to                                                                        
InnoDB: http://dev.mysql.com/doc/refman/5.6/en/forcing-innodb-recovery.html                                                         
InnoDB: about forcing recovery.

How to repeat:
my.cnf

[mysqld]
gdb
innodb_file_per_table = true
innodb_data_file_path = ibdata1:100M:autoextend
innodb_flush_method = O_DIRECT
innodb_log_buffer_size = 256M

innodb_flush_log_at_trx_commit = 1
innodb_buffer_pool_size = 300G
innodb_buffer_pool_instances=8

innodb_log_file_size = 4G
innodb_log_files_in_group = 2
#####plugin options
innodb_read_io_threads = 16
innodb_write_io_threads = 16
innodb_io_capacity = 20000
innodb_io_capacity_max = 40000
#not innodb options (fixed)
port = 3306
back_log = 50
max_connections = 2000
max_prepared_stmt_count=500000
max_connect_errors = 10
table_open_cache = 2048
max_allowed_packet = 16M
binlog_cache_size = 16M
max_heap_table_size = 64M
sort_buffer_size = 4M
join_buffer_size = 4M
thread_cache_size = 1000
query_cache_size = 0
query_cache_type = 0
ft_min_word_len = 4
#default_table_type = InnoDB
thread_stack = 192K
tmp_table_size = 64M

server-id = 10
#*** MyISAM Specific options
key_buffer_size = 8M
read_buffer_size = 1M
read_rnd_buffer_size = 4M
bulk_insert_buffer_size = 8M
myisam_sort_buffer_size = 8M
myisam_max_sort_file_size = 10G
#myisam_max_extra_sort_file_size = 10G
myisam_repair_threads = 1
myisam_recover
#socket=/var/lib/mysql/mysql.sock
user=root
skip-grant-tables

additional parameters:
--defaults-file=conf/tpcc/tpccw2500-innodb.mysql.cnf --datadir=/mnt/fiob/data --basedir=/usr/local/mysql-5.6.7-rc-linux2.6-x86_64  -
-innodb_buffer_pool_size=125G --innodb_log_file_size=4G --innodb_doublewrite=1 --innodb_buffer_pool_instances=8 --core  --log-error=mysqld.err                             
environment: LD_PRELOAD=/data/opt/alexey.s/numactl-2.0.7/libnuma.so /data/opt/alexey.s/numactl-2.0.7/numactl --interleave=all /usr/b
in/env LD_PRELOAD=/usr/local/lib/libjemalloc.so  

tpcc parameters:
tpcc_start  -c 32 -h 127.0.0.1 -P 3306 -d tpccw2500 -u root -p "" -w 2500 -r 10 -l 4000 -i 1
[12 Oct 2012 13:46] Inaam Rana
Vadim,

Thanks for reporting this. We have a fix for this going through internal review. The bug is a race in changed doublewrite buffer code.
[29 Dec 2012 16:33] Erlend Dahl
This is a duplicate of an internally filed bug that was fixed in 5.6.9.