使用innobackupex基于从库搭建级联从库及一两从

江枫思渺然 提交于 2020-04-02 18:27:57

使用innobackupex基于从库搭建mysql主从架构

现有的架构是一主一从,版本为Mysql5.6.37。实施要求是:利用从库,搭建第二个从库,版本为5.7.21

主库:192.168.1.21 

从库:192.168.1.22

从库2:192.168.1.23

1、备份从库1
[root@mysql02 ~]./innobackupex --user=root --password="xxxxxx" --slave-info --safe-slave-backup --parallel=3 /data/backup/

#参数slave-info 用于GTID
#复制my.cnf文件到新从库
[root@mysql02 ~]$scp /etc/my.cnf 192.168.1.23:/data/backup
2、主库授予从库2复制账户
master@MySQL> grant replication slave,replication client on *.* to repl@'192.168.1.23' identified by 'repl';
3、从库2:进行恢复
[root@mysql03 ~]$./innobackupex --user=root --password="xxxxxx" --apply-log --use-memory=2G /data/backup/2018-08-10_17-29-01

./innobackupex --defaults-file=/etc/my.cnf  --user=root --password="xxxxxx"  --socket=/data/mysql_33061/run/mysql.sock --copy-back  /data/backup/2018-08-10_17-29-01

或者:
[root@mysql03 ~]$mv /data/backup/2018-08-10_17-29-01 /data/mysql_3306

4、准备从库2配置文件my.cnf,根据需要修改相应参数,这里的修改如下,
skip-slave-start
datadir = /data/mysql_3306
port = 3306
server_id = 330623
socket = /data/mysql_3306/mysql.sock
pid-file=/data/mysql_3306/mysql3307.pid
log_error=/data/mysql_3306/mysql_error.log


5、启动从库2
# chown -R mysql:mysql /data/backup
# mysqld_safe --defaults-file=/etc/my.cnf &

6、运行mysql时,提示Table ‘performance_schema.session_variables’ doesn’t exist
./mysql_upgrade -hlocalhost -u root -p --force -P3306 --socket=/data/mysql_3306/mysql.sock
重启数据库
7、change master
mysql> system more /log/recover/xtrabackup_slave_info
CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000658', MASTER_LOG_POS=925384099
mysql> CHANGE MASTER TO
-> MASTER_HOST='192.168.1.22',
-> MASTER_USER='repl',
-> MASTER_PASSWORD='repl',
-> MASTER_PORT=3306,
-> MASTER_LOG_FILE='mysql-bin.000658',
-> MASTER_LOG_POS=925384099;
Query OK, 0 rows affected, 2 warnings (0.31 sec)
mysql> start slave;
在从库2,start slave时报:error:
ERROR 1794 (HY000): Slave is not configured or failed to initialize properly. You must at least set --server-id to enable either a master or a slave. Additional error messages can be found in the MySQL error log.
经分析,需要:
删除5张表,并重新导入脚本
use mysql
drop table slave_master_info;
drop table slave_relay_log_info;
drop table slave_worker_info;
drop table innodb_index_stats;
drop table innodb_table_stats;
检查表数据,为空表,重建表
source /backup/mysql_system_tables.sql

(2)重新启动数据库
mysql> start slave;
mysql> show slave status\G;

 一主一从一级联,搭建完成。

 

未开启GTID情况下,从 一主一从一级联,调整为一主两从

1:在从库一:停掉SQL线程、IO线程,并记录下此时从库1 的master status

mysql>stop slave;

mysql> show master status;
+------------------+-----------+--------------+------------------+-------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |
+------------------+-----------+--------------+------------------+-------------------+
| mysql-bin.000658| 925384099| | | |
+------------------+-----------+--------------+------------------+-------------------+
1 row in set (0.00 sec)

2:在从库2,show slave status;观察Master_Log_File  Exec_Master_Log_Pos与从库1中show master status;一致时,在从库1:show slave status\G;记录下Master_Log_File  Exec_Master_Log_Pos

3:然后在从库2,即指向主库:

mysql> CHANGE MASTER TO
-> MASTER_HOST='192.168.1.21', 
-> MASTER_USER='repl', 
-> MASTER_PASSWORD='repl',
-> MASTER_PORT=3306,
-> MASTER_LOG_FILE='mysql-bin.000100',
-> MASTER_LOG_POS=384099;
Query OK, 0 rows affected, 2 warnings (0.31 sec)
mysql> start slave;

4:开启从库1的SQL线程、IO线程

此时,一主两从搭建成功;

 

易学教程内所有资源均来自网络或用户发布的内容,如有违反法律规定的内容欢迎反馈
该文章没有解决你所遇到的问题?点击提问,说说你的问题,让更多的人一起探讨吧!