Bug #27924 FLUSH LOGS crashes database
Submitted: 18 Apr 2007 10:36 Modified: 21 May 2007 10:27
Reporter: Ronald Klop Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Server: General Severity:S2 (Serious)
Version:5.0.24-Debian_1~bpo.1-log OS:Linux (Debian 3.1 i386)
Assigned to: CPU Architecture:Any

[18 Apr 2007 10:36] Ronald Klop
Description:
When we issue a FLUSH LOGS command the server crashes.
The stacktrace offers this.
# resolve_stack_dump -s /tmp/mysqld.sym -n /tmp/mysql.stack 
0x81ba8e4 handle_segfault + 660
0xffffe420 _end + -141362896
0x2 (?)
0x8231956 _ZN9MYSQL_LOG22purge_logs_before_dateEl + 86
0x8232b3f _ZN9MYSQL_LOG16rotate_and_purgeEj + 207
0x81d8f73 _Z20reload_acl_and_cacheP3THDmP13st_table_listPb + 771
0x81d2205 _Z21mysql_execute_commandP3THD + 13157
0x81d7910 _Z11mysql_parseP3THDPcj + 304
0x81cd827 _Z16dispatch_command19enum_server_commandP3THDPcj + 1159
0x81cd34d _Z10do_commandP3THD + 141
0x81cc7e2 handle_one_connection + 466
0xb7f7ab63 _end + -1349861261
0xb7df218a _end + -1351469414

This is my config.
[client]
port            = 3306
socket          = /var/run/mysqld/mysqld.sock

[mysqld_safe]
socket          = /var/run/mysqld/mysqld.sock
nice            = 0

[mysqld]
#read-only
skip-slave-start
skip-external-locking
slave-load-tmpdir       = /data/mysql/tmp
#log-slave-updates

# only keep 1 day since this is a slave not a master.
# also we copy the DB to the test machine from here.
expire-logs-days        = 1

user            = mysql
server-id       = 14
#log-bin                = rwlinux14-bin
socket          = /var/run/mysqld/mysqld.sock
pid-file        = /tmp/mysql.pid

port            = 3306
basedir         = /usr
datadir         = /data/mysql/data
tmpdir          = /tmp
language        = /usr/share/mysql/english

old_passwords   = 1

key_buffer              = 200M
max_allowed_packet      = 32M
sort_buffer             = 2M
record_buffer           = 1M
myisam_sort_buffer_size = 64M
table_cache             = 2048
thread_stack            = 256K
thread_cache_size       = 64
query_cache_size        = 104857600 
query_cache_type        = 1
log-slow-queries        = /var/log/mysql/mysql-slow.log
long_query_time         = 1

max_connections         = 1000
open_files_limit        = 8192
max_connect_errors      = 500
connect_timeout         = 10
thread_concurrency      = 8

skip-bdb

#Aan zetten na conversie naar UTF8?
#character-set-server   = utf8
#default-character-set  = utf8
#collation-server       = utf8_general_ci

default-table-type      = innodb

innodb_data_home_dir            = /data/mysql/data
innodb_data_file_path           = ibdata1:4000M:autoextend
innodb_log_group_home_dir       = /data/mysql/data
innodb_log_arch_dir             = /data/mysql/data
innodb_log_archive              = 0
innodb_mirrored_log_groups      = 1
innodb_log_files_in_group       = 2
innodb_log_file_size            = 50M
innodb_log_buffer_size          = 25M
innodb_flush_log_at_trx_commit  = 1
innodb_buffer_pool_size         = 1000M
innodb_additional_mem_pool_size = 200M
innodb_file_io_threads          = 16
innodb_lock_wait_timeout        = 50

[mysqldump]
quick
quote-names
max_allowed_packet      = 16M

[mysql]
#no-auto-rehash # faster start of mysql but no tab completition

[isamchk]
key_buffer              = 16M

How to repeat:
Issue a FLUSH LOGS command.
[18 Apr 2007 10:39] Ronald Klop
This might be the same as this issue in the changelog of 5.0.33.
- FLUSH LOGS or mysqladmin flush-logs caused a server crash if the binary log was not open. (Bug#17733)
[18 Apr 2007 10:42] Valeriy Kravchuk
Thank you for a problem report. Please, try to repeat with a newer version, 5.0.37/5.0.38, and inform about the results.
[18 May 2007 23:00] Bugs System
No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".
[21 May 2007 10:27] Ronald Klop
In my comment on 18 Apr 12:39 I mentioned another issue report. This is a duplicate of it.