Mysql+Keepalived实现双主热备方式
目录
- 一、Mysql安装、配置和安全启动
- 二、 两台Mysql配置双主
- 三、安装Keepalived
- 四、配置Keepalived
我们通常说的双机热备是指两台机器都在运行,但并不是两台机器都同时在提供服务。当提供服务的一台出现故障的时候,另外一台会马上自动接管并且提供服务,而且切换的时间非常短
MySQL双主复制,即互为Master-Slave(默认只使用一台MasterA负责数据写入,另一台MasterB备用),可以实现数据库服务器的热备,但是一台Master宕机后不能实现动态切换。
使用Keepalived,可以通过虚拟IP,实现双主对外的统一接口以及自动检查、失败切换机制,从而实现MySQL数据库的高可用方案。
一、Mysql安装、配置和安全启动
把mysql-5.7.28-linux-glibc2.12-x86_64.tar.gz上传到/usr/local/src/
1、mysql安装
cd /usr/local/src/ tar -zxvf mysql-5.7.28-linux-glibc2.12-x86_64.tar.gz mkdir -p /usr/local/mysql/ mv mysql-5.7.28-linux-glibc2.12-x86_64 /usr/local/mysql/mysql-5.7.28
2、创建mysql用户组和用户
cd /usr/local groupadd mysql useradd -r -g mysql -s /bin/false mysql chown -R mysql:mysql mysql
3、初始化数据
cd /usr/local/mysql/mysql-5.7.28 bin/mysqld --initialize --user=mysql --basedir=/usr/local/mysql/mysql-5.7.28/ --datadir=/data1/mysql/data --lc_messages_dir=/usr/local/mysql/mysql-5.7.28/share --lc_messages=en_US bin/mysql_ssl_rsa_setup --datadir=/data1/mysql/data
4、修改配置
vi /etc/my.cnf basedir=/usr/local/mysql/mysql-5.7.28 datadir=/data1/mysql/data
5、配置环境变量
#编辑环境变量 .bash_profile表示当前用户个人配置 只对当前用户生效 vi ~/.bash_profile #将mysql的bin目录配置到环境变量中 PATH=$PATH:$HOME/bin:/usr/local/mysql/mysql-5.7.28/bin export PATH #使配置修改后立即生效 source ~/.bash_profile
6、修改密码、配置远程访问和安全启动
#启动前先设置免密登录 vi /etc/my.cnf #添加如下配置并保存 跳过权限验证 skip-grant-tables=1 #安全启动 cd /usr/local/mysql/mysql-5.7.28/bin ./mysqld_safe --user=mysql & #修改密码 #进入mysql mysql use mysql update user set authentication_string=password("123456") where user="root"; #刷新权限表 flush privileges; #配置远程访问 use mysql #允许所有机器访问root用户 %代表所有机器 update user set host='%' where user='root'; #授权所有机器的root用户拥有所有数据库的所有权限 远程访问密码root %代表所有机器 grant all privileges on *.* to root@'%' identified by 'root'; #刷新权限表 flush privileges; #退出mysql exit #安全关闭 ./mysqladmin -uroot -p shutdown #删除免密登录配置 vi /etc/my.cnf #删除skip-grant-tables=1并保存 #安全启动 ./mysqld_safe --user=mysql &
二、 两台Mysql配置双主
1、创建主从同步replication用户
2、授权
192.168.0.101
#授权IP192.168.0.102的从数据库replication用户拥有所有数据库的同步权限 远程访问密码replication grant replication slave on *.* to 'replication'@'192.168.0.102' identified by 'replication'; #刷新权限表 flush privileges;
192.168.0.102
#授权IP192.168.0.101的从数据库replication用户拥有所有数据库的同步权限 远程访问密码replication grant replication slave on *.* to 'replication'@'192.168.0.101' identified by 'replication'; #刷新权限表 flush privileges;
3、添加配置文件
[mysqld] basedir=/usr/local/mysql/mysql-5.7.28 datadir=/data1/mysql/data server-id = 101 skip-external-locking log_bin = mysql-binlog binlog_do_db = sample binlog-ignore-db = mysql binlog-ignore-db = performance_schema binlog-ignore-db = information_schema binlog_group_commit_sync_delay = 100000 # 单位微妙 binlog_group_commit_sync_no_delay_count = 20 # 单位每组事物数量 #设置复制类型 slave-parallel-type=LOGICAL_CLOCK #设置并行数量 slave-parallel-workers=4 #主主需要加的部分 replicate-do-db = sample log-slave-updates=true #sync_binlog=1 auto_increment_offset=1 auto_increment_increment=2 innodb_buffer_pool_size = 12G innodb_buffer_pool_instances = 8 innodb_buffer_pool_chunk_size = 128M innodb_log_file_size = 256M innodb_log_buffer_size = 32M innodb_flush_method = O_DIRECT innodb_lock_wait_timeout = 120 slow_query_log = ON slow_query_log_file = /usr/local/mysql/log/slow.log long_query_time = 5 sql_mode=STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION secure_file_priv=/usr/local/mysqlcsvfiles/ expire_logs_days=7 max_binlog_size=1024M !includedir /etc/my.cnf.d
4、添加之后
mkdir -p /usr/local/mysql/log mkdir -p /usr/local/mysqlcsvfiles
5、安全关闭
./mysqladmin -uroot -p shutdown
6、安全启动
./mysqld_safe --user=mysql &
7、进入mysql执行
#进入mysql执行 show master status\G; #分别记录主的binlog位置,在101的角度上102为主,在102的角度上,101为主 File:mysql-bin.000001 Position: 154 #设置同步关系(两个主都要设置) #其中master_log_file和master_log_pos要登录到对方机器执行show master status\G;去查看 stop slave; (在192.168.0.101) change master to master_host='192.168.0.102',master_user='replication',master_password='replication',master_log_file='mysql-bin.000001',master_log_pos=154; (在192.168.0.102) change master to master_host='192.168.0.101',master_user='replication',master_password='replication',master_log_file='mysql-bin.000001',master_log_pos=154; #分别启动slave start slave;
三、安装Keepalived
1、安装编译依赖包
yum install -y gcc openssl-devel libnl libnl-devel libnfnetlink-devel
2、安装Keepalived
#解压keepalived压缩包放到/usr/local/并重命名为keepalived cd /usr/local/src tar -zxvf keepalived-2.0.7.tar.gz mv keepalived-2.0.7 /usr/local/keepalived #配置keepalived 得到一个Makefile的文件夹 #--prefix:keepalived安装目录 #--sysconf:keepalived的核心配置文件,必须要在/etc目录下面,改为其他位置会导致启动不了,不配置在该目录下的话,启动keepalived时日志文件里面会报错,显示找不到/etc这个文件夹 cd /usr/local/keepalived ./configure --prefix=/usr/local/keepalived/ --sysconf /etc #编译和安装keepalived make && make install #创建keepalived软链接 /usr/sbin/如果存在keepalived就先删除 ln -s /usr/local/keepalived/sbin/keepalived /usr/sbin/ #复制keepalived脚本文件到/etc/init.d/目录下 cd keepalived cp /usr/local/keepalived/etc/init.d/keepalived /etc/init.d/ #设置Keepalived开机自启动 chkconfig --add keepalived chkconfig keepalived on #启动keepalived service keepalived start
四、配置Keepalived
1、编辑keepalived配置文件
vi /etc/keepalived/keepalived.conf
2、主服务器添加对应配置文件
global_defs { router_id LVS_LEVEL1 #主服务器名称 } vrrp_script check_run { script "/etc/keepalived/mysql_check.sh" interval 5 #5秒执行一次脚本 } vrrp_instance VI_1 { state MASTER #主服务器 interface eth0 #承载VIP地址到物理接口 virtual_router_id 51 #虚拟路由器ID号,每个热播组保持一致 priority 100 #优先级,数值越大优先级越高 advert_int 1 #检查间隔,默认为1s authentication { #认证信息,每个热播组保持一致 auth_type PASS #认证类型 auth_pass 1111 #密码字串 } virtual_ipaddress { 192.168.0.144 #VIP地址(内网地址) } track_script { check_run } }
3、备份服务器添加对应配置文件
global_defs { router_id LVS_LEVEL2 #备份服务器名称 } vrrp_script check_run { script "/etc/keepalived/mysql_check.sh" interval 5 #5秒执行一次脚本 } vrrp_instance VI_1 { state BACKUP #备份服务器 interface eth0 #承载VIP地址到物理接口 virtual_router_id 51 #虚拟路由器ID号,每个热播组保持一致 priority 50 #优先级,数值越大优先级越高 advert_int 1 #检查间隔,默认为1s authentication { #认证信息,每个热播组保持一致 auth_type PASS #认证类型 auth_pass 1111 #密码字串 } virtual_ipaddress { 192.168.0.144 #VIP地址(和主服务器设置一样) } track_script { check_run } }
附mysql_check.sh
#!/bin/bash /usr/bin/mysql -uroot -p'你自己的数据库密码' -e "show status" &>/dev/null if [ $? -ne 0 ] ;then systemctl stop keepalived fi
chmod +x /etc/keepalived/mysql_check.sh
4、重启keepalived
以上为个人经验,希望能给大家一个参考,也希望大家多多支持我们。
赞 (0)