23.1、mysql高可用业务需求:


23、mysql高可用实践_mysql

23.2、mysql高可用架构图(单主热备模式):


23、mysql高可用实践_主从同步_02


23、mysql高可用实践_主从同步_03

23.6、部署drbd、heartbeat、mysql的主从同步:

1、部署heartbeat:参考"21、部署herartbeat"文档;

2、部署drbd:参考"22、部署drbd"文档;

3、heartbeat和drbd整合:参考"22.17、heartbeat和drbd整合"文档;

3、部署mysql同步:参考"14、mysql主从复制实战"、"3、mysql的多实例配置"文档;

23.7、配置细节说明:

停止主和从的heartbeat和drbd服务;

1、heartbeat:

(1)修改maste-db和slave-db的/etc/ha.d/haresource配置文件:

(2)master-db IPaddr::172.16.1.26/24/eth1 drbddisk::data Filesystem::/dev/drbd0::/master-db::ext4 mysql

2、drbd:

(1)启动master-db和slave-db上的drbd服务;

(2)使用'drbdadm primary data'将master-db端的drdb服务置主;

(3)手动挂载master-db端的'/dev/drbd0'设备到'/master-db/'目录上;

3、mysql:

(1)在master-db和slave-db相同的路径下安装mysql软件;

(2)master-db端在/master-db/目录下面初始化mysql并指定data目录为/master-db/,为mysql创建登录密码,启动mysql测试;

此时在master-db的操作都同步到了slave-db端的/dev/drbd0管理的/dev/sdb1系统分区下面;slave-db端只要保证有/master-db/目录即可;

[root@master-db ~]#tree /master-db/

/master-db/

├── 3306

│   ├── my.cnf

│   └── mysql

│ └── data

└── lost+found

2 directories, 3 files

(3)在master-db端和slave-db端将mysql脚本拷贝到/etc/ha.d/resource.d/目录下面,方便heartbeat管理mysql,要加执行的权限;

(4)为了方便测试在slave-db端安装端口为3307的mysql多实例进行测;

23.8、测试:

停止master-db和slave-db上的mysql、heartbeat、drbd服务;

1、启动顺序:

(1)master-db:

1)drbdadm up data

2)/etc/init.d/heartbeat start

(2)slave-db:

1)drbdadm up data

2)/etc/init.d/heartbeat start

3)/data/3307/mysql start

2、结果:

(1)master-db:

1)df -h

Filesystem Size Used Avail Use% Mounted on

/dev/drbd0 4.5G 1.4G 2.9G 33% /master-db

2)cat /proc/drbd

0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r-----

ns:12 nr:0 dw:4 dr:817 al:1 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:0

3)ip addr

3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:5c:f0:d0 brd ff:ff:ff:ff:ff:ff

inet 172.16.1.16/24 brd 172.16.1.255 scope global eth1

inet 172.16.1.26/24 brd 172.16.1.255 scope global secondary eth1

inet6 fe80::20c:29ff:fe5c:f0d0/64 scope link

valid_lft forever preferred_lft forever

4)ss -tunlp | grep 3306

tcp LISTEN 0 600 *:3306 *:* user

(2)slave-db:

1)ip addr、df -h、ss -tunlp | grep 3306无结果:

2)cat /proc/drbd

0: cs:Connected ro:Secondary/Primary ds:UpToDate/UpToDate C r-----

ns:12 nr:0 dw:4 dr:817 al:1 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:0

3)查看3307主从同步结果:

show slave status\G;

3、master-db和slave-db进行切换:

(1)在master-db上执行'/usr/share/heartbeat/hb_standy'命令将资源交给slave-db端,

此时结果和上面的结果想法;此时会发现slave-db端的3307 mysql实例在show slave status\G;状态时

会出现正在尝试连接主端,等待60s就可以了,在等待的60s期间由于vip已经漂移到了salve-db端,磁盘

已经挂载(drbd此端为主,对端为从),mysql启动,所以用户还是可以往数据库中写数据的;

(2)在主端执行'/usr/share/heartbeat/hb_takeover local'master-db从slave-db接管回资源;

4、关闭顺序:

(1)slave-db:

/data/3307/mysql stop

/etc/init.d/heartbeat stop

drbdadm down data

(2)master-db:

/etc/init.d/heartbeat stop

drbdadm down data