MySQL灾备切换

1.1 配置mysql主从

主库IP:192.168.8.62

从库IP:192.168.8.65

主库IP:192.168.8.62 操作

    mysql -uroot -p
	mysql> grant replication slave on *.* to tongbu@'192.168.8.65' identified by '123456';

	mysql> show master status \G
	*************************** 1. row ***************************
				 File: mysql-bin.000005
			 Position: 331
	#拿到File 和Position
	#锁表
	flush tables with read lock;   锁表

1.2 从库IP:192.168.8.65 操作

    mysql> change master to master_host='192.168.8.62',master_user='tongbu',master_password='123456',master_log_file='mysql-bin.000005',master_log_pos=331;
	mysql> start  slave;
	mysql> show slave status \G
		#内容有这两给 YES 
		Slave_IO_Running:Yes
		Slave_SQL_Running:yes
		
	# 在主库上运行 表锁释放命令 192.168.8.62
	mysql> unlock tables
	# 完成了 主从配置

1.3,配置my.cof

   ##############  主mysql 的my.cof  ##############
	[mysqld]
	datadir=/var/lib/mysql
	socket=/var/lib/mysql/mysql.sock
	character-set-server=utf8
	max_connections=1000
	interactive_timeout=31536000
	wait_timeout=31536000
	lower_case_table_names=1
	symbolic-links=0
	log-bin=mysql-bin
	server-id = 1               # 配置主数据库的ID
	replicate-ignore-db=mysql   # 配置不同步的库
	replicate-ignore-db=test
	replicate-ignore-db=performance_schema
	replicate-ignore-db=information_schema
	# Disabling symbolic-links is recommended to prevent assorted security risks
	symbolic-links=0
	# Recommended in standard MySQL setup
	sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES 

	##################  从mysql 的my.cof  ########################

	[mysqld]
	datadir=/var/lib/mysql
	socket=/var/lib/mysql/mysql.sock
	character-set-server=utf8
	max_connections=1000
	interactive_timeout=31536000
	wait_timeout=31536000
	lower_case_table_names=1
	symbolic-links=0
	log-bin=mysql-bin
	server-id = 3               # 配置从数据库的ID
	replicate-ignore-db=mysql   # 配置不同步的库
	replicate-ignore-db=test
	replicate-ignore-db=performance_schema
	replicate-ignore-db=information_schema
	# Disabling symbolic-links is recommended to prevent assorted security risks
	symbolic-links=0
	# Recommended in standard MySQL setup
	sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES 

	#######################################################

2. mysql切换灾备服务器

2.1 登陆mysql 在salve执行：

	mysql> stop slave；
	mysql> reset master；

2.2删除mysql目录下

relay-log.info 
master.info
rm -f /var/lib/mysql/*.info

2.3 如果my.cof配置上存在log-slave-updates read_only 等一定注释掉它

2.4重启mysql

service  mysql  restart

redis 切换

2.5 登陆redis

redis-cli 
AUTH  uhfugqpowzffhDWQc20z-pUadl&2d1@Dkzqlo/adfwQQlo+AdjcdUyakLjFM$YUaxOasGDBeyTail5Is9Ush+0aXPiU
#密码命令
slaveof NO ONE
#停止redis 主从

修改redis.conf 注释掉它
#slaveof 192.168.1.55 #master IP

2.6 启动tomcat 和zk

3.1mysql binlog恢复数据

编辑再还原数据

#导出 2019-01-14  15:12分到2019-01-14 15:16:00 的数据
mysqlbinlog -uroot -p'123456' --start-datetime="2019-01-14 15:12:00" --stop-datetime="2019-01-14 15:16:00" /var/lib/mysql/mysql-bin.000003 > mysql_restore.sql
mysql -uroot -p < /tmp/mysql_restore.sql
#直接还原数据
mysqlbinlog -uroot -p --start-datetime="2019-01-14 15:12:00" --stop-datetime="2019-01-14 15:16:00" /var/lib/mysql/mysql-bin.000003 | mysql -u root -p'123456'

​

mysql 备份方式恢复数据

全量备份数据脚本

MYUSER=script_user
MYSQL_PWD=123123
mysql -e "show databases;" -uroot -p'123456' | egrep -vi "Data|_schema|mysql|test|job|order"| \
xargs mysqldump -u$MYUSER -p$MYSQL_PWD --databases |gzip> /www/backup/mysql_$(date +%F)_dump.all.sql.gz
2)还原数据
gzip -d mysql_dump.all.sql.gz
mysql  -u$MYUSER -p$MYSQL_PWD  < mysql_dump.all.sql

redis 切换

1. 登陆redis
   redis-cli
   AUTH uhfugqpowzffhDWQc20z-pUadl&2d1@Dkzqlo/adfwQQlo+AdjcdUyakLjFM$YUaxOasGDBeyTail5Is9Ush+0aXPiU

2. 在salve执行
   slaveof 192.168.8.62 6379

3. 修改redis.conf 注释掉它

   slaveof 192.168.8.62 #master IP

附 mysql 操作

锁表

flush tables with read lock;

表锁释放

unlock tables;

刷新 logs

flush logs

刷新权限

flush privileges;

运行查看状态

show master status \G