异步复制(默认复制)
主数据库在完成后会立即返回给客户端,不管从数据库有没有同步上,这也就导致如果主数据库在返回后down掉时,从数据库有可能没有数据,导致从数据库替代后,没有上一条数据
-
每个事务在更新完成数据之前,master会在二进制日志(binary log,binlog)中记录sql语句,slave开启i/o线程,将sql语句写入自己的日志,并重新识别
-
所有的从数据库版本都要高于主数据库
-
状态:show slave status;
实验
MASTER
两台虚拟机10,11
关闭防火墙三件套
yum -y install ntp #同步时间
vim /etc/ntp.conf
server 127.127.1.0
fudge 127.127.1.0 stratum 8
systemctl enable ntpd #开机自启
systemctl start ntopd
yum -y install ntpdate #时间模块
/user/sbin/ntpdate 192.168.200.11 #同步从数据库时间
vim /etc/my.cnf #修改配置文件
server-id=1
log-bin=mysql-binlog
log-slave-updates=true
systemctl start mariadb #启动
mysql
#打开主数据库授权方便从连接
grant replication slave on *.* to 'myslave'@'192.168.200.%' identified by '123123';
flush privileges; #刷新授权表
show master status; #查看master状态
slave
mysql -uroot -p123123 -h 192.168.200.10 #测试连接主数据库
vim /etc/my.cnf #修改配置文件
server-id=2
relay-log=relay-log-bin
relay-log-index=slaver-relay-bin.index
systemctl restat mariadb
stop slave; #停止slave
#授权用户
change master to master_host='192.168.200.10',master_user='myslave',master_password='123123',master_log_file='mysql-binlog.0.000003',master_log_pos=522;
start slave;
show slave statusG
配置解析
change master to
master_host #主服务器IP地址
master_user #主服务器授权用户
master_password #主服务器授权密码
master_log_file #主服务器二进制日志文件名
master_log_pos #日志文件开始位置
- 如果连接后没有同步数据:先暂停slave set global_slave_skip_counter=1 在启动
mysql 主主+keepalived
思路
- 两台mysql都可以读写,互为主备,默认使用一台masterA,负责数据的写入,另一台备用
- masterA是masterB的主库,B又是A的主库
- 使用Keepalived的VIP作为条件连接
缺点
- masterB会一直处于空闲的状态
- 从库要等masterB完成之后再去复制,会有一定时间的延迟
实验
三台机器10,11,12(10,12互为主)
关闭防火墙三件套
vim /etc/my.cnf #配置文件
mysql
grant replication slave on *.* to 'repl'@'192.168.200.11' identified by '123123'; #授权
show master status; #查看点
+---------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+---------------------+----------+--------------+------------------+
| mysql-binlog.000005 | 548 | | |
+---------------------+----------+--------------+------------------+
##配置11
vim /etc/my.cnf
grant replication slave on *.* to 'repl'@'192.168.200.11' identified by '123123';
MariaDB [(none)]> show master status;
+---------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+---------------------+----------+--------------+------------------+
| mysql-binlog.000001 | 786 | | |
+---------------------+----------+--------------+------------------+
stop slave;
change master to master_host='192.168.200.11',master_port=3306,master_user='repl',master_password='123123',master_log_file='mysql-binlog.000001',master_log_pos=786;
show slave status G
#masterB
grant replication slave on *.* to 'repl'@'192.168.200.14' identified by '123123';
MariaDB [(none)]> show master status;
stop slave;
change master to master_host='192.168.200.14',master_port=3306,master_user='repl',master_password='123123',master_log_file='mysql-binlog.000001',master_log_pos=786;
show slave status G
##当两台的id,sql线程为yes时成功
测试
masterA配置代码
[mysqld]
server-id=1
log-bin=mysql-binlog #打开二进制功能
log-slave-updates=true
max_binlog_size=1024M #binlog单文件最大值
auto_increment_offset = 1
auto_increment_increment = 2 #奇数ID
replicate-ignore-db = information_schema #忽略不同步主从的数据库
replicate-ignore-db = performance_schema
replicate-ignore-db = test
replicate-ignore-db = mysql
max_connections = 3000
max_connect_errors = 30
skip-character-set-client-handshake #忽略应用程序想要设置的其他字符集
init-connect='SET NAMES utf8' #连接时执行的SQL
character-set-server=utf8 #服务端默认字符集
wait_timeout=1800 #请求的最大连接时间
interactive_timeout=1800 #和上一参数同时修改才会生效
sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES #sql模式
relay-log=relay-log-bin #开启中继日志
relay-log-index=slave-relay-bin.index
masterB配置代码
[mysqld]
server-id=2
log-bin=mysql-binlog
log-slave-updates=true
max_binlog_size=1024M
auto_increment_offset = 2
auto_increment_increment = 2
replicate-ignore-db = information_schema
replicate-ignore-db = performance_schema
replicate-ignore-db = test
replicate-ignore-db = mysql
max_connections = 3000
max_connect_errors = 30
skip-character-set-client-handshake
init-connect='SET NAMES utf8'
character-set-server=utf8
wait_timeout=1800
interactive_timeout=1800
sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES
relay-log=relay-log-bin
relay-log-index=slave-relay-bin.index
Keepalived高可用
yum -y install keepalived #下载Keepalived
vim /etc/keeplived/keeplived.conf #配置文件
vim /opt/mysql.sh #写入脚本
chmod +X /opt/mysql.sh #给脚本授权
systemctl start keepalived
ip a | grep 254
#masterB同时写入mysql.sh与my.cnf配置文件的优先级改为99,id改为2
#测试:重启mariadb,Keepalived
slave:mysql -uroot -p123123 -h 192.168.200.254
关闭masterA的mysql查看masterB的254
Keepalived配置文件
! Configuration File for keepalived
global_defs {
router_id LVS_MASTER-A
}
vrrp_script mysql {
script "/opt/mysql.sh"
interval 2
weight -5
fall 2
rise 1
}
vrrp_instance VI_1 {
state BACKUP
interface ens32
virtual_router_id 51
priority 100
nopreempt
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
track_script {
mysql
}
virtual_ipaddress {
192.168.200.254
}
}
mysql.sh脚本
#!/bin/bash
counter=$(netstat -na|grep 'LISTEN'|grep '3306' |wc-l)
if ['${counter}' -eq 0 ]; then
systemctl stop keepalived
fi