Bug #2032 mysqld is ignoring binlog sizes set in the my.cnf file
Submitted: 6 Dec 2003 13:58 Modified: 7 Dec 2003 23:15
Reporter: Chris Waskowich Email Updates:
Status: Duplicate Impact on me:
None 
Category:MySQL Server: Replication Severity:S2 (Serious)
Version:4.0.16 OS:MacOS (Mac OS X Server 10.3)
Assigned to: CPU Architecture:Any

[6 Dec 2003 13:58] Chris Waskowich
Description:
On my slave server, with a self-complied version of mysql, the mysqld seems to be ignoring the 
values set for the binlog sizes.  I have them set in the my.cnf file, and I in the sql prompt:

mysql> show variables like "%log%size%";
+------------------------+------------+
| Variable_name          | Value      |
+------------------------+------------+
| binlog_cache_size      | 32768      |
| innodb_log_buffer_size | 1048576    |
| innodb_log_file_size   | 5242880    |
| max_binlog_cache_size  | 4294967295 |
| max_binlog_size        | 134217728  |
| max_relay_log_size     | 134217728  |
+------------------------+------------+
6 rows in set (0.00 sec)

So, it seems to be reading the data from my.cnf file, however, in implementation, it does not seem 
to be working.

The relay-bin logs never grow beyond 32K and if I enable the 'log-bin' option, the mysqld will 
crash.  Per the documentation at http://www.mysql.com/doc/en/
Replication_Implementation_Details.html the relay log files get cycled only when a flush logs 
command is issued or when the log has exceed the 'max_relay_log_size' value.  So, either the 
master is issuing a 'flush logs' or mysqld is not working correctly.  I believe it is the later as when 
another slave (with an older version of the OS) is using the same version of mysql and the same 
my.cnf file it correctly interprets the values set in the my.cnf file.

Here is my my.cnf file.  It is the only my.cnf file located on my file system and I have indeed verified 
that the mysqld is reading from it.

[client]
port            = 3306
socket          = /tmp/mysql.sock

[mysqld]
port            = 3306
socket          = /tmp/mysql.sock

skip-locking

key_buffer = 128M
max_allowed_packet = 1M
thread_stack = 64K
table_cache = 32
sort_buffer = 64K
net_buffer_length = 2K

max_relay_log_size = 128M
max_binlog_size = 128M

server-id = 2

master-host = <removed>
master-user = <removed>
master-password = <removed>
master-port = 3306
master-connect-retry = 60

replicate-do-db = db1
replicate-do-db = db2
replicate-do-db = db3

#log-bin
skip-bdb
skip-innodb

[mysqldump]
quick
max_allowed_packet=16M

[mysql]
no-auto-rehash

[isamchk]
key_buffer=8M
sort_buffer=8M

[myisamchk]
key_buffer=8M
sort_buffer=8M

[mysqlhotcopy]
interactive-timeout

How to repeat:
Can't stop it from repeating.

Suggested fix:
None.
[7 Dec 2003 23:15] Alexander Keremidarski
Please do not submit the same bug more than once. An existing
bug report already describes this very problem. Even if you feel
that your issue is somewhat different, the resolution is likely
to be the same. Because of this, we hope you add your comments
to the original bug instead.

Thank you for your interest in MySQL.

Additional info:

Same as Bug #2030 http://bugs.mysql.com/bug.php?id=2030