1、什麼是GTID？
1、全局唯一，一個事務對應一個GTID 2、替代傳統的binlog+pos複製；使用master_auto_position=1自動匹配GTID斷點進行復制 3、MySQL5.6開始支持 4、在傳統的主從複製中，slave端不用開啓binlog；但是在GTID主從複製中，必須開啓binlog 5、slave端在接受master的binlog時，會校驗GTID值 6、爲了保證主從數據的一致性，多線程同時執行一個GTID
2、組成
Master_UUID:序列號
舉例：
ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-5
ceb0ca3d-8366-11e8-ad2b-000c298b7c9a其實就是master的uuid值；1-5是序列號，每次一個事務完成都會自增1，也就是說下一次爲1-6。
3、工作原理
1、master更新數據時，會在事務前產生GTID，一同記錄到binlog日誌中。 2、slave端的i/o 線程將變更的binlog，寫入到本地的relay log中。 3、sql線程從relay log中獲取GTID，然後對比slave端的binlog是否有記錄。 4、如果有記錄，說明該GTID的事務已經執行，slave會忽略。 5、如果沒有記錄，slave就會從relay log中執行該GTID的事務，並記錄到binlog。 6、在解析過程中會判斷是否有主鍵，如果沒有就用二級索引，如果沒有就用全部掃描
4、GTID主從配置
版本：MySQL5.7
配置master
vim /etc/my.cnf [client] socket=/usr/local/mysql/mysql.sock [mysqld] basedir=/usr/local/mysql datadir=/usr/local/mysql/data user=mysql pid-file=/usr/local/mysql/data/mysqld.pid log-error=/usr/local/mysql/data/mysql.err socket=/usr/local/mysql/mysql.sock port=3306 server-id=1 gtid-mode=ON enforce-gtid-consistency=ON server-id=1 binlog_format=row log-bin=/usr/local/mysql/data/mysql-bin systemctl restart mysqld firewall-cmd --add-port=3306/tcp --permanent firewall-cmd --reload
配置slave
vim /etc/my.cnf [client] socket=/usr/local/mysql/mysql.sock [mysqld] basedir=/usr/local/mysql datadir=/usr/local/mysql/data user=mysql pid-file=/usr/local/mysql/data/mysqld.pid log-error=/usr/local/mysql/data/mysql.err socket=/usr/local/mysql/mysql.sock port=3306 server-id=2 gtid-mode=ON enforce-gtid-consistency=ON server-id=2 binlog_format=ROW log-bin=/usr/local/mysql/data/mysql-bin log_slave_updates=ON skip-slave-start=1 systemctl restart mysqld firewall-cmd --add-port=3306/tcp --permanent firewall-cmd --reload
master授權配置
mysql -uroot -p mysql> grant replication slave on *.* to 'rep'@'10.0.0.%' identified by '123'; mysql> flush privileges;
slave配置同步
mysql -uroot -p mysql> change master to master_host='10.0.0.132', master_user='rep',master_password='123',master_port=3306,master_auto_position=1; mysql> start slave;
查看slave的狀態
mysql> show slave status\G; *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 10.0.0.132 Master_User: rep Master_Port: 3306 Connect_Retry: 60 Master_Log_File: mysql-bin.000003 Read_Master_Log_Pos: 635 Relay_Log_File: slave-relay-bin.000005 Relay_Log_Pos: 848 Relay_Master_Log_File: mysql-bin.000003 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: 635 Relay_Log_Space: 1308 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: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a Master_Info_File: /usr/local/mysql/data/master.info SQL_Delay: 0 SQL_Remaining_Delay: NULL Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates Master_Retry_Count: 86400 Master_Bind: Last_IO_Error_Timestamp: Last_SQL_Error_Timestamp: Master_SSL_Crl: Master_SSL_Crlpath: Retrieved_Gtid_Set: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-4 Executed_Gtid_Set: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-4 Auto_Position: 1 Replicate_Rewrite_DB: Channel_Name: Master_TLS_Version: 1 row in set (0.00 sec)
出現這兩個yes表示同步成功
通過slave的狀態信息，可以看到GTID的值、Matser_UUID等信息
查看master狀態
mysql> show master status; +------------------+----------+--------------+------------------+------------------------------------------+ | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set | +------------------+----------+--------------+------------------+------------------------------------------+ | mysql-bin.000003 | 635 | | | ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-4 | +------------------+----------+--------------+------------------+------------------------------------------+ 1 row in set (0.00 sec)
注意對比slave端，Executed_Gtid_Set的值應該是一樣的。
5、驗證主從
master上
mysql> create database test01; Query OK, 1 row affected (0.00 sec) mysql> show master status; +------------------+----------+--------------+------------------+------------------------------------------+ | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set | +------------------+----------+--------------+------------------+------------------------------------------+ | mysql-bin.000003 | 800 | | | ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-5 | +------------------+----------+--------------+------------------+------------------------------------------+ 1 row in set (0.00 sec)
slave上
mysql> show databases; +--------------------+ | Database | +--------------------+ | information_schema | | mysql | | performance_schema | | sys | | test01 | +--------------------+ 5 rows in set (0.07 sec) mysql> show slave status\G; *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 10.0.0.132 Master_User: rep Master_Port: 3306 Connect_Retry: 60 Master_Log_File: mysql-bin.000003 Read_Master_Log_Pos: 800 Relay_Log_File: slave-relay-bin.000005 Relay_Log_Pos: 1013 Relay_Master_Log_File: mysql-bin.000003 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: 800 Relay_Log_Space: 1473 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: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a Master_Info_File: /usr/local/mysql/data/master.info SQL_Delay: 0 SQL_Remaining_Delay: NULL Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates Master_Retry_Count: 86400 Master_Bind: Last_IO_Error_Timestamp: Last_SQL_Error_Timestamp: Master_SSL_Crl: Master_SSL_Crlpath: Retrieved_Gtid_Set: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-5 Executed_Gtid_Set: ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-5 Auto_Position: 1 Replicate_Rewrite_DB: Channel_Name: Master_TLS_Version: 1 row in set (0.00 sec)
需要注意的是，GTID的值在完成一次事務後，變成了ceb0ca3d-8366-11e8-ad2b-000c298b7c9a:1-5（自增1）
6、排障
思路
a、確保master開放3306端口
b、最好關閉selinux
c、master上授權同步，slave上change master命令指定master的信息不要寫錯
d、UUID問題
如果你出現了上圖所示的問題，表示你的master和slave的UUID是一樣的，一般這種情況多出現於克隆虛擬機
解決辦法：
找到slave上的MySQL數據目錄下的auto.cnf文件（這個文件其實是自動生成的mysql服務器的UUID值），將它刪除，然後重啓MySQL，然後MySQL會重新生成一個UUID。然後停掉slave，重新開啓就可以了（我的mysql的數據目錄是在/usr/local/mysql/data下，詳情查看my.cnf配置文件）
cd /usr/local/mysql/data rm -f auto.cnf systemctl restart mysql
[root@slave data]# cat auto.cnf [auto] server-uuid=020c7f26-be57-11e8-8e2d-000c29b63bad
通過cat命令查看該文件，發現UUID已經改變
mysql -uroot -p mysql> stop slave; mysql> start slave;
e、總結
排障過程中，注意需要停掉slave，做完修改之後在開啓，否則你的修改可能是不會生效的。