MySQL5.6 Replication主從復制（讀寫分離) 配置完整版

MySQL5.6主從復制(讀寫分離)教程

1、MySQL5.6開始主從復制有兩種方式：

基於日志（binlog）；
基於GTID（全局事務標示符）。

需要注意的是：GTID方式不支持臨時表！所以如果你的業務系統要用到臨時表的話就不要考慮這種方式了，至少目前最新版本MySQL5.6.12的GTID復制還是不支持臨時表的。

所以本教程主要是告訴大家如何通過日志（binlog）方式做主從復制！

2、MySQL官方提供的MySQL Replication教程：

http://dev.mysql.com/doc/refman/5.6/en/replication.html

第一步：准備工作

主服務器: 192.168.1.100
從服務器: 192.168.1.101

MySQL軟件版本：

MySQL-server-advanced-5.6.18-1.el6.x86_64.rpm
MySQL-cient-advanced-5.6.18-1.el6.x86_64.rpm

第二步：在主服務器和從服務器上安裝MySQL數據庫軟件

安裝方法，請參見 http://www.jb51.net/article/82542.htm

MySQL數據庫軟件安裝完成後，不要急著做mysql啟動操作。建議把mysql初始化生成的/usr/my.cnf
(如果是從源文件編譯安裝時，路徑應該是在/usr/local/mysql/mysql.cnf)刪除，然後把優化好的mysql
配置文件my.cnf放到/etc下。
第三步：修改主數據庫的配置文件/usr/my.cnf
復制代碼 代碼如下:
[mysqld]
 
server-id=1
log-bin=mysqlmaster-bin.log
sync_binlog=1

innodb_buffer_pool_size=512M
innodb_flush_log_at_trx_commit=1
 
sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
 
lower_case_table_names=1
log_bin_trust_function_creators=1

第四步：修改從數據庫配置文件/usr/my.cnf
 
復制代碼 代碼如下:
server-id=2
log-bin=mysqlslave-bin.log
sync_binlog=1
innodb_buffer_pool_size=512M
innodb_flush_log_at_trx_commit=1
 

sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
lower_case_table_names=1
log_bin_trust_function_creators=1

第五步：在主數據庫和從數據庫服務器上分別執行以下命令重新啟動主數據庫和從數據庫
復制代碼 代碼如下:
[root@master ~]# service mysql restart
[root@slave ~]# service mysql restart

第六步：在主數據庫上創建用於主從復制的賬戶

復制代碼 代碼如下:
[root@master ~]# mysql -uroot -p
mysql> GRANT REPLICATION SLAVE ON *.* TO 'repl'@'192.168.1.101' IDENTIFIED BY '111111';
Query OK, 0 rows affected (0.00 sec)

注意：以上命令中的IP地址，是從數據庫服務器的IP地址。

第七步：主數據庫鎖表(禁止再插入數據以獲取主數據庫的的二進制日志坐標)
mysql> flush tables with read lock;
Query OK, 0 rows affected (0.00 sec)
 

第八步：查看主數據庫的狀態(並記錄下File字段和Position字段的值，在配置從服務器時有用到)

mysql> show master status;
+------------------------+----------+--------------+------------------+-------------------+
| File                  | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |
+------------------------+----------+--------------+------------------+-------------------+
| mysqlmaster-bin.000004 |      327 |              |                  |                  |
+------------------------+----------+--------------+------------------+-------------------+
1 row in set (0.00 sec)

第九步：創建主數據庫的快照文件
[root@master ~]# cd /usr/bin/
# ./mysqldump -uroot -p -h127.0.0.1 -P3306 --all-databases --triggers --routines --events >>/mnt/windows/all.sql
上面命令中的紅色部分，是一個共享目錄，這個目錄可以同時被主數據庫服務器和從數據庫服務器訪問到。
如果沒有這樣的共享目錄，可以將all.sql放在其它任何目錄下，然後使用scp命令復制到遠程從數據庫服務器的某個目錄中
這條命令的執行時間根據數據量的不同，會有所不同，如果主數據庫的數據量很大，可能需要很長時間，那麼在這種情況下，就最好在晚上沒有業務的時候進行這個操作，否則第七步中的鎖表操作會對業務系統造成很大的影響
第十步：解鎖主數據庫的鎖表操作
[root@master ~]# mysql -uroot -p    (本命令在主數據庫服務器上執行)
mysql> unlock tables;
Query OK, 0 rows affected (0.00 sec)
第十一步：在從數據庫服務器上導入第七步創建的快照文件到從數據庫中
[root@slave ~]# mysql -uroot -p -h127.0.0.1 -P3306 < /mnt/windows/all.sql
第十二步：在從數據庫服務器上設置主數據庫服務器向從數據庫服務器同步
[root@slave ~]# mysql -uroot -p
mysql> change master to master_host = '192.168.1.100',master_user='repl',master_password='111111',master_log_file='mysqlmaster-bin.000004',master_log_pos=327;
注意：紅色部分的值，是在第八步中查出來的，這裡不能弄錯了
第十三步：啟動從數據庫復制線程
mysql> start slave;
Query OK, 0 rows affected (0.01 sec)
第十四步：查詢從數據庫的復制線程狀態

復制代碼 代碼如下:
mysql> show slave status \G
*************************** 1. row ***************************
              Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.1.100
                  Master_User: repl
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: mysqlmaster-bin.000004
          Read_Master_Log_Pos: 327
              Relay_Log_File: slave-relay-bin.000002
                Relay_Log_Pos: 289
        Relay_Master_Log_File: mysqlmaster-bin.000004
          Slave_IO_Running: Yes
        Slave_SQL_Running: Yes
              Replicate_Do_DB:
          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: 327
              Relay_Log_Space: 462
              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:
  Replicate_Ignore_Server_Ids:
            Master_Server_Id: 1
                  Master_UUID: 2e5e1b22-f0a9-11e3-bbac-000c297799e0
            Master_Info_File: /var/lib/mysql/master.info
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it
          Master_Retry_Count: 86400
                  Master_Bind:
      Last_IO_Error_Timestamp:
    Last_SQL_Error_Timestamp:
              Master_SSL_Crl:
          Master_SSL_Crlpath:
          Retrieved_Gtid_Set:
            Executed_Gtid_Set:
                Auto_Position: 0
1 row in set (0.00 sec)

如果Slave_IO_Running和Slave_SQL_Running兩項都為yes，就表示主從復制配置成功了.
下面可以開始測試配置是否成功了，首先在主數據庫的test數據庫中新建一張表，然後插入幾條數據，然後到從數據庫看看是否同步過來了。
注意：當從數據庫有大量的查詢時，可以暫時將從數據庫的復制線程關閉掉，等查詢量降下來了，再打開，這樣也不會丟失數據。

附：一個優化好後的主數據庫配置文件和從數據配置文件內容如下：

復制代碼 代碼如下:
# For advice on how to change settings please see
# http://dev.mysql.com/doc/refman/5.6/en/server-configuration-defaults.html
[client]
port=3306
socket=/usr/local/mysql/mysql.sock
default-character-set=utf8
 
[mysqld]
sync_binlog=1
server-id=1
port=3306
socket=/usr/local/mysql/mysql.sock
pid-file=/home/mysql/temp/my3306.pid
user=mysql
datadir=/home/mysql/data
tmpdir=/home/mysql/temp/
log-bin=/home/mysql/data/mysqlmaster-bin
log-error=/home/mysql/logs/error.log
slow_query_log_file=/home/mysql/logs/slow.log
binlog_format=mixed
slow_query_log
long_query_time=10
wait_timeout=31536000
interactive_timeout=31536000
max_connections=500
max_user_connections=490
max_connect_errors=2
character_set_server=utf8
skip-external-locking
key_buffer_size = 128M
max_allowed_packet = 5M
table_open_cache = 512
sort_buffer_size = 2M
read_buffer_size = 2M
read_rnd_buffer_size = 8M
myisam_sort_buffer_size = 64M
thread_cache_size = 8
query_cache_size = 32M
# Try number of CPU's*2 for thread_concurrency
thread_concurrency = 4
binlog-ignore-db=mysql
binlog-ignore-db=information_schema
replicate_ignore_db=mysql
replicate_ignore_db=information_schema
expire-logs-days=10
skip-slave-start
skip-name-resolve
lower_case_table_names=1
log_bin_trust_function_creators=1
 
# InnoDB
innodb_data_home_dir=/home/mysql/data
innodb_log_group_home_dir=/home/mysql/logs
innodb_data_file_path=ibdata1:128M:autoextend
innodb_buffer_pool_size=2G
innodb_log_file_size=10M
innodb_log_buffer_size=8M
innodb_lock_wait_timeout=50
innodb_file_per_table
innodb_flush_log_at_trx_commit=1
 
#sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
 

 
 
一個優化好的從數據庫的配置文件如下：

復制代碼 代碼如下:
# For advice on how to change settings please see
# http://dev.mysql.com/doc/refman/5.6/en/server-configuration-defaults.html
[client]
port=3306
socket=/usr/local/mysql/mysql.sock
default-character-set=utf8
 
[mysqld]
sync_binlog=1
server-id=2
port=3306
socket=/usr/local/mysql/mysql.sock
pid-file=/home/mysql/temp/my3306.pid
user=mysql
datadir=/home/mysql/data
tmpdir=/home/mysql/temp/
log-bin=/home/mysql/data/mysqlslave-bin
log-error=/home/mysql/logs/error.log
slow_query_log_file=/home/mysql/logs/slow.log
binlog_format=mixed
slow_query_log
long_query_time=10
wait_timeout=31536000
interactive_timeout=31536000
max_connections=500
max_user_connections=490
max_connect_errors=2
character_set_server=utf8
skip-external-locking
key_buffer_size = 128M
max_allowed_packet = 5M
table_open_cache = 512
sort_buffer_size = 2M
read_buffer_size = 2M
read_rnd_buffer_size = 8M
myisam_sort_buffer_size = 64M
thread_cache_size = 8
query_cache_size = 32M
# Try number of CPU's*2 for thread_concurrency
thread_concurrency = 4
binlog-ignore-db=mysql
binlog-ignore-db=information_schema
replicate_ignore_db=mysql
replicate_ignore_db=information_schema
expire-logs-days=10
#skip-slave-start
skip-name-resolve
lower_case_table_names=1
log_bin_trust_function_creators=1
 
# InnoDB
innodb_data_home_dir=/home/mysql/data
innodb_log_group_home_dir=/home/mysql/logs
innodb_data_file_path=ibdata1:128M:autoextend
innodb_buffer_pool_size=2G
innodb_log_file_size=10M
innodb_log_buffer_size=8M
innodb_lock_wait_timeout=50
innodb_file_per_table
innodb_flush_log_at_trx_commit=1
 
#sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
sql_mode=STRICT_TRANS_TABLES,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION,NO_AUTO_VALUE_ON_ZERO
 
[mysqldump]
quick
max_allowed_packet = 16M
 
[mysql]
no-auto-rehash
 
[myisamchk]
key_buffer_size = 256K
sort_buffer_size = 256K
read_buffer = 256K
write_buffer = 256K
 
[mysqlhotcopy]
interactive-timeout
 
sql_mode=STRICT_TRANS_TABLES,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION,NO_AUTO_VALUE_ON_ZERO
 
[mysqldump]
quick
max_allowed_packet = 16M
 
[mysql]
no-auto-rehash
 
[myisamchk]
key_buffer_size = 256K
sort_buffer_size = 256K
read_buffer = 256K
write_buffer = 256K
 
[mysqlhotcopy]
interactive-timeout