準備工作
Mysql 單機多實例詳解
What is the GTID of the replication
實驗環境
Os: CentOS 6.X
Mysql: 5.6 單機多實例 [3306,3307]
Hostname: lab.suzf.net
場景一：新機器 無數據
對於GTID的配置，主要修改配置文件中與GTID特性相關的幾個重要參數
[mysqld_multi] mysqld = /usr/local/mysql/bin/mysqld_safe mysqladmin = /usr/local/mysql/bin/mysqladmin log = /usr/local/mysql/data/log/mysqld_multi.log user = root ## Used for stopping the server via mysqladmin # master [mysqld3306] socket = /usr/local/mysql/data/run/mysqld_3306.sock port = 3306 pid-file = /usr/local/mysql/data/run/mysqld_3306.pid datadir = /usr/local/mysql/data/mysql_3306 # set for GTID Replication server_id = 3232237894 gtid_mode = on enforce_gtid_consistency = on #強制gtid一致性，開啓後對於特定create table不被支持 # binlog log-bin = /usr/local/mysql/data/binlogs/master-binlog log-slave-updates = 1 binlog_format = row #強烈建議，其他格式可能造成數據不一致 # rely log skip_slave_start=1 # slave [mysqld3307] socket = /usr/local/mysql/data/run/mysqld_3307.sock port = 3307 pid-file = /usr/local/mysql/data/run/mysqld_3307.pid datadir = /usr/local/mysql/data/mysql_3307 # set for GTID Replication server_id = 3232237866 gtid_mode = on enforce_gtid_consistency = on # bin log log-bin = /usr/local/mysql/data/binlogs/slave-binlog log-slave-updates = 1 binlog_format = row # rely log skip_slave_start=1 [client] default-character-set = utf8
重啓mysql
/etc/init.d/mysqld_multi.server restart
Mysql 多實例啓動腳本
cat /etc/init.d/mysqld_multi.server
#!/bin/sh
#
# A simple startup script for mysqld_multi by Tim Smith and Jani Tolonen.
# This script assumes that my.cnf file exists either in /etc/my.cnf or
# /root/.my.cnf and has groups [mysqld_multi] and [mysqldN]. See the
# mysqld_multi documentation for detailed instructions.
#
# This script can be used as /etc/init.d/mysql.server
#
# Comments to support chkconfig on RedHat Linux
# chkconfig: 2345 64 36
# description: A very fast and reliable SQL database engine.
#
# Version 1.0
#
basedir=/usr/local/mysql
bindir=/usr/local/mysql/bin
conf=$basedir/data/etc/my.cnf
export PATH=$PATH:$bindir
if test -x $bindir/mysqld_multi
then
mysqld_multi="$bindir/mysqld_multi";
else
echo "Can't execute $bindir/mysqld_multi from dir $basedir";
exit;
fi
case "$1" in
'start' )
"$mysqld_multi" --defaults-extra-file=$conf start $2
;;
'stop' )
"$mysqld_multi" --defaults-extra-file=$conf stop $2
;;
'report' )
"$mysqld_multi" --defaults-extra-file=$conf report $2
;;
'restart' )
"$mysqld_multi" --defaults-extra-file=$conf stop $2
"$mysqld_multi" --defaults-extra-file=$conf start $2
;;
*)
echo "Usage: $0 {start|stop|report|restart}" >&2
;;
esacMaster OPS
mysql -uroot -S /usr/local/mysql/data/run/mysqld_3306.sock mysql> GRANT REPLICATION SLAVE ON *.* TO 'repluser'@'localhost' IDENTIFIED BY 'suzf.net666'; Query OK, 0 rows affected (0.00 sec)
Slave OPS
mysql -uroot -S /usr/local/mysql/data/run/mysqld_3307.sock mysql> CHANGE MASTER TO -> MASTER_HOST='localhost', -> MASTER_USER='repluser', -> MASTER_PASSWORD='suzf.net666', -> MASTER_PORT=3306, -> MASTER_AUTO_POSITION = 1; Query OK, 0 rows affected, 2 warnings (0.04 sec) mysql> start slave; Query OK, 0 rows affected (0.19 sec)
驗證
場景二：移除 Replication
首先關閉 SLAVE
mysql> STOP SLAVE; Query OK, 0 rows affected (0.00 sec)
使用ALL 參數 重置 SLAVE (MySQL >= 5.6.7):
mysql> RESET SLAVE ALL; Query OK, 0 rows affected (0.02 sec)
此時 mysql replication 已經消失了
mysql> SHOW SLAVE STATUS\G Empty set (0.00 sec)
如果你希望這個SLAVE節點繼續做一個新的MASTER的slave 節點，那麼請設置 gtid_purged,你應該發出一下命令（即使 gtid_puerged 看起來是空的，你也應該重置一下MASTER.）
mysql> reset master; Query OK, 0 rows affected (0.11 sec) mysql> show variables like '%gtid%'; +--------------------------+-----------+ | Variable_name | Value | +--------------------------+-----------+ | enforce_gtid_consistency | ON | | gtid_executed | | | gtid_mode | ON | | gtid_next | AUTOMATIC | | gtid_owned | | | gtid_purged | | +--------------------------+-----------+ 6 rows in set (0.00 sec)
備註
我很確定我已經刪除了 rely-binary-logs. (relay_log_basename variable)
場景三：拷貝數據 新建 GTID slave
Using mysqldump
在 slave 幾點初始化數據庫
$ ./scripts/mysql_install_db --datadir=/usr/local/mysql/data/mysql_3307
爲了更簡單使用 mysqldump, 是將下列參數加入配置文件
[mysqldump] quick max_allowed_packet = 16M port = 3306 socket = /usr/local/mysql/data/run/mysqld_3306.sock
之後使用下面命令導出 Master 節點的 Cherry 數據庫
$ mysqldump --defaults-file=/usr/local/mysql/data/etc/my.cnf --user=root --password=`cat ~mysql/.root_password` --single-transaction --databases Cherry > backup.sql Warning: Using a password on the command line interface can be insecure. Warning: A partial dump from a server that has GTIDs will by default include the GTIDs of all transactions, even those that changed suppressed parts of the database. If you don't want to restore GTIDs, pass --set-gtid-purged=OFF. To make a complete dump, pass --all-databases --triggers --routines --events.
備註
–defaults-file 參數必須放在第一個位置, 否則你將會入到一個我認爲已經修復的 bug.
GTID 的訣竅在這裏，當然它是通過 mysqldump 來實現的
$grep -i gtid 、/tmp/backup.sql -- GTID state at the beginning of the backup SET @@GLOBAL.GTID_PURGED='770d3753-c6e4-11e2-8e78-080027d93e15:1-8';
之後將dump 文件傳送到 slave 節點，然後導入它；
mysql> source /tmp/backup.sql;
我最初得到了下面的錯誤：
ERROR 1840 (HY000): @@GLOBAL.GTID_PURGED can only be set when @@GLOBAL.GTID_EXECUTED is empty.
但是我的 gtid_executed 參數是空的:
mysql> show variables like '%gtid%'; +--------------------------+------------------------------------------+ | Variable_name | Value | +--------------------------+------------------------------------------+ | enforce_gtid_consistency | ON | | gtid_executed | | | gtid_mode | ON | | gtid_next | AUTOMATIC | | gtid_owned | | | gtid_purged | 770d3753-c6e4-11e2-8e78-080027d93e15:1-6 | +--------------------------+------------------------------------------+ 6 rows in set (0.00 sec)
錯誤消息顯示衝突 一個bug已經打開。實際上，gtid_purged也必須是空的，以便能夠設置它。官方文檔中引用的唯一解決方法是使用下面的命令：
mysql> DROP DATABASE replicationdb; Query OK, 1 row affected (0.07 sec) mysql> RESET MASTER; Query OK, 0 rows affected (0.06 sec) mysql> SHOW VARIABLES LIKE '%gtid%'; +--------------------------+-----------+ | Variable_name | Value | +--------------------------+-----------+ | enforce_gtid_consistency | ON | | gtid_executed | | | gtid_mode | ON | | gtid_next | AUTOMATIC | | gtid_owned | | | gtid_purged | | +--------------------------+-----------+ 6 rows in set (0.01 sec)
一旦完成你可以從新導入備份文件，如果最後沒有錯誤出現，那麼現在可以設置 replication 了。
mysql> change master to master_host='lab.suzf.net', master_port=3306, master_user='repluser', master_password='suzf.net666', master_auto_position=1; Query OK, 0 rows affected, 2 warnings (0.16 sec) mysql> start slave; Query OK, 0 rows affected (0.03 sec) mysql> SHOW SLAVE STATUS\G *************************** 1. row *************************** Slave_IO_Running: Yes Slave_SQL_Running: Yes ... ... Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it
如果使用的是mysql 5.5 可能會用到
change master to master_host='lab.suzf.net', master_port=3306, master_user='repluser', master_password='suzf.net666', master_log_file='mysql-bin.000006', master_log_pos=1026;
從哪裏可以獲得 master_log_file 和 log_position（在你導入數據時請不要將 master 設置爲 read-only ）
mysql> FLUSH TABLES WITH READ LOCK; mysql> SHOW MASTER STATUS;
Using XtraBackup
準備工作
$mkdir /opt/data/backup/mysql -p $ cat 3306.cnf [mysqld] socket = /usr/local/mysql/data/run/mysqld_3306.sock port = 3306 pid-file = /usr/local/mysql/data/run/mysqld_3306.pid datadir = /usr/local/mysql/data/mysql_3306 # set for GTID Replication server_id = 3232237894 gtid_mode = on enforce_gtid_consistency = on # binlog log-bin = /usr/local/mysql/data/binlogs/master-binlog log-slave-updates = 1 binlog_format = row # rely log skip_slave_start=1 [client] default-character-set = utf8 $ cat 3307.cnf [mysqld] socket = /usr/local/mysql/data/run/mysqld_3307.sock port = 3307 pid-file = /usr/local/mysql/data/run/mysqld_3307.pid datadir = /usr/local/mysql/data/mysql_3307 # set for GTID Replication server_id = 3232237866 gtid_mode = on enforce_gtid_consistency = on # bin log log-bin = /usr/local/mysql/data/binlogs/slave-binlog log-slave-updates = 1 binlog_format = row # rely log skip_slave_start=1 [client] default-character-set = utf8
我們將使用下面命令完整數據庫備份：
$innobackupex --defaults-file=/opt/data/backup/mysql/3306.cnf --user=root --socket=/usr/local/mysql/data/run/mysqld_3306.sock /tmp
這裏要注意的重要的文件是在xtrabackup_binlog_info
$ cat /tmp/2016-04-05_15-26-40/xtrabackup_binlog_info master-binlog.000006 191 81ee72fe-c957-11e5-ae70-0800272aa66e:1-12
在 Slave 節點恢復數據
$ innobackupex --defaults-file=/opt/data/backup/mysql/3307.cnf --user=root --copy-back /tmp/2016-04-05_15-26-40
備註
Original data directory must be empty!
即 Slave 節點數據目錄必須爲空
權限修復 啓動實例
$ chown -R mysql:mysql /usr/local/mysql/data/mysql_3307 $ /etc/init.d/mysqld_multi.server start 3307 $ /etc/init.d/mysqld_multi.server report Reporting MySQL servers MySQL server from group: mysqld3306 is running MySQL server from group: mysqld3307 is running
你需要先設置 gtid_purged 然後再創建 replication，在 Percona 文檔 中有說明
mysql> SET GLOBAL gtid_purged="81ee72fe-c957-11e5-ae70-0800272aa66e:1-12"; ERROR 1840 (HY000): @@GLOBAL.GTID_PURGED can only be set when @@GLOBAL.GTID_EXECUTED is empty. # 出現上面個錯誤 執行下面操作 mysql> reset master; Query OK, 0 rows affected (0.02 sec) mysql> SET GLOBAL gtid_purged="81ee72fe-c957-11e5-ae70-0800272aa66e:1-12"; Query OK, 0 rows affected (0.05 sec) mysql> CHANGE MASTER TO -> MASTER_HOST='localhost', -> MASTER_USER='repluser', -> MASTER_PASSWORD='suzf.net666', -> MASTER_PORT=3306, -> MASTER_AUTO_POSITION = 1; Query OK, 0 rows affected, 2 warnings (0.06 sec) mysql> SHOW WARNINGS; +-------+------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | Level | Code | Message | +-------+------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | Note | 1759 | Sending passwords in plain text without SSL/TLS is extremely insecure. | | Note | 1760 | Storing MySQL user name or password information in the master info repository is not secure and is therefore not recommended. Please consider using the USER and PASSWORD connection options for START SLAVE; see the 'START SLAVE Syntax' in the MySQL Manual for more information. | +-------+------+--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ 2 rows in set (0.00 sec)
在經典的複製中（MySQL的5.5及以下）命令應該是這樣的：
change master to master_host='lab.suzf.net', master_port=3306, master_user='repluser', master_password='suzf.net666', master_log_file='mysql-bin.000006', master_log_pos=1026;
啓動 slave 查看狀態
mysql> start slave; Query OK, 0 rows affected (0.01 sec) mysql> show slave status\G Slave_IO_Running: Yes Slave_SQL_Running: Yes Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it Retrieved_Gtid_Set: 81ee72fe-c957-11e5-ae70-0800272aa66e:13 Executed_Gtid_Set: 81ee72fe-c957-11e5-ae70-0800272aa66e:1-13 Auto_Position: 1
場景四：經典複製 –> GTID Replication
a. 按場景一中描述配置參數文件
b. 所有服務器設置global.read_only參數，等待主從服務器同步完畢；
mysql> SET @@global.read_only = ON;
c. 依次重啓主從服務器；
d. 使用change master 更新主從配置；
mysql> CHANGE MASTER TO
> MASTER_HOST = host,
> MASTER_PORT = port,
> MASTER_USER = user,
> MASTER_PASSWORD = password,
> MASTER_AUTO_POSITION = 1;
e. 從庫開啓複製
mysql> START SLAVE;
f. 驗證主從複製
FAQ
Skip counter with GTID
mysql> show slave status\G *************************** 1. row *************************** Retrieved_Gtid_Set: 770d3753-c6e4-11e2-8e78-080027d93e15:1-9 Executed_Gtid_Set: 770d3753-c6e4-11e2-8e78-080027d93e15:1-7, 97e23f6a-cc52-11e2-b1e1-08002776e125:1-2 Auto_Position: 1 1 row in set (0.00 sec)
從上面我們可以猜測到 事務執行到 事務ID 8 可能出現錯誤，接下來我們需要注入相同的空事務ID。
mysql> stop slave; Query OK, 0 rows affected (0.00 sec) mysql> show variables like '%gtid%'; +--------------------------+-----------+ | Variable_name | Value | +--------------------------+-----------+ | enforce_gtid_consistency | ON | | gtid_executed | | | gtid_mode | ON | | gtid_next | AUTOMATIC | | gtid_owned | | | gtid_purged | | +--------------------------+-----------+ 6 rows in set (0.00 sec) mysql> set gtid_next='770d3753-c6e4-11e2-8e78-080027d93e15:8'; Query OK, 0 rows affected (0.00 sec) mysql> begin; Query OK, 0 rows affected (0.00 sec) mysql> commit; Query OK, 0 rows affected (0.00 sec) mysql> set gtid_next='AUTOMATIC'; Query OK, 0 rows affected (0.00 sec) mysql> start slave; Query OK, 0 rows affected (0.00 sec)
Reference : Mysql manual
至此 完 
License：Attribution-NonCommercial-NoDerivatives 4.0 International
本文出自 Suzf Blog。 如未註明，均爲 SUZF.NET 原創。
轉載請註明出處：http://suzf.net/thread-0401-696.html