Question

我在Red Hat Linux上托管MySQL Ver 14.14，Distrib 5.1.16的两台不同服务器上进行了Master-to-master复制设置。当我重新启动其中一个服务器时，从站无法启动。当我执行/ var / lib / mysql目录的列表时，我注意到master.info文件被截断为零长度，这导致MySQL认为没有设置复制。

这是服务器1的my.cnf：

[client]
port = 3306
socket = /var/lib/mysql/mysql.sock

[mysqldump]
quick
max_allowed_packet = 16M

[mysql]
no-auto-rehash

[isamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[myisamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[mysqlhotcopy]
interactive-timeout

[mysqld]
port = 3306
socket = /var/lib/mysql/mysql.sock
key_buffer  = 16M
max_allowed_packet = 1M
table_cache = 64
sort_buffer_size = 512K
net_buffer_length = 8K
myisam_sort_buffer_size = 8M
log-bin = mysql-bin
relay-log = mysqld-relay-bin
relay-log-index = mysqld-relay-bin.index
server-id = 101
binlog-format = STATEMENT
replicate-do-db = foo
replicate-do-db = bar
binlog-do-db = foo
binlog-do-db = bar
auto_increment_increment = 2
auto_increment_offset = 1
master-connect-retry = 2
sync_binlog = 1
log-error = mysqld.log
log-warnings = 2
wait_timeout = 31536000
expire_logs_days = 45

这是服务器2的my.cnf：

[client]
port = 3306
socket = /var/lib/mysql/mysql.sock

[mysqldump]
quick
max_allowed_packet = 16M

[mysql]
no-auto-rehash

[isamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[myisamchk]
key_buffer = 20M
sort_buffer_size = 20M
read_buffer = 2M
write_buffer = 2M

[mysqlhotcopy]
interactive-timeout

[mysqld]
port = 3306
socket = /var/lib/mysql/mysql.sock
key_buffer  = 16M
max_allowed_packet = 1M
table_cache = 64
sort_buffer_size = 512K
net_buffer_length = 8K
myisam_sort_buffer_size = 8M
log-bin = mysql-bin
relay-log = mysqld-relay-bin
relay-log-index = mysqld-relay-bin.index
server-id = 102
binlog-format = STATEMENT
replicate-do-db = foo
replicate-do-db = bar
binlog-do-db = foo
binlog-do-db = bar
auto_increment_increment = 2
auto_increment_offset = 2
master-connect-retry = 2
sync_binlog = 1
log-error = mysqld.log
log-warnings = 2
wait_timeout = 31536000
expire_logs_days = 45

我在每台服务器上设置了奴隶，如下所示：

STOP SLAVE ; RESET SLAVE ; CHANGE MASTER TO MASTER_HOST='other_sys', MASTER_USER='repl', MASTER_PASSWORD='super_secret_password', MASTER_PORT=3306, MASTER_LOG_FILE='mysql-bin.000001', MASTER_LOG_POS=14048 ;

MASTER_LOG_FILE和MASTER_LOG_POS在上面的示例中只是任意的。执行此操作后，我得到以下MySQL从属状态：

mysql> show slave status \G
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: other_sys
                  Master_User: repl
                  Master_Port: 3306
                Connect_Retry: 2
              Master_Log_File: mysql-bin.000002
          Read_Master_Log_Pos: 14048
               Relay_Log_File: mysqld-relay-bin.000002
                Relay_Log_Pos: 251
        Relay_Master_Log_File: mysql-bin.000002
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
              Replicate_Do_DB: foo,bar
          Replicate_Ignore_DB:
           Replicate_Do_Table:
       Replicate_Ignore_Table:
      Replicate_Wild_Do_Table:
  Replicate_Wild_Ignore_Table:
                   Last_Errno: 0
                   Last_Error:
                 Skip_Counter: 0
          Exec_Master_Log_Pos: 14048
              Relay_Log_Space: 407
              Until_Condition: None
               Until_Log_File:
                Until_Log_Pos: 0
           Master_SSL_Allowed: No
           Master_SSL_CA_File:
           Master_SSL_CA_Path:
              Master_SSL_Cert:
            Master_SSL_Cipher:
               Master_SSL_Key:
        Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
                Last_IO_Errno: 0
                Last_IO_Error:
               Last_SQL_Errno: 0
               Last_SQL_Error:
1 row in set (0.00 sec)

然后，我使用命令reboot重新启动服务器。重新启动后，MySQL从站有时会从它停止的位置自动启动。其他时候，MySQL slave根本不会启动，我注意到show slave status \G返回一个空集，而/var/lib/mysql/master.info文件被截断为零长度。这几乎就像Linux在重启时没有将文件缓冲区刷新到inode，因此从站信息不会被保存。

在配置奴隶时我是否遗漏了什么？

Answer 1

万一有人想知道我是如何解决的：

问题是Linux文件缓冲。我执行了上述相同的步骤，除了我在调用log4j.category.org.springframework.web.servlet.mvc.method.annotation.RequestMappingHandlerMapping=TRACE之前调用了sync，并且它在100次尝试中100％工作。没有reboot，它在我的测试中失败了10次。我不知道为什么Linux在关机时没有同步文件缓存，但是在发出sync之前调用sync或等待一分钟已经解决了问题。

master.info文件不断被截断

1 个答案: