Build a High‑Availability MySQL Cluster with Corosync, Pacemaker & DRBD

Implementation topology diagram:

Experiment Procedure:

先对两主机安装上corosync和pacemaker

两主机安装drbd服务，并且设置好drbd的组设备

选择drbd主节点上，进行数据库的初始化

进行资源的定义和配置

测试

Pre‑experiment preparation:

1. Time must be synchronized

使用ntp服务器

2. Nodes must be able to resolve each other's names

解析节点名称
/etc/host
集群中使用的主机名为`uname -n`表示的主机名；

3. Ping node

仅偶数节点才需要；

4. SSH key authentication for communication

Experiment Steps:

Install corosync and pacemaker:

yum install corosync pacemaker -y

Configure corosync and let pacemaker run as a plugin:

vim /etc/corosync/corosync.conf

compatibility: whitetank

totem {
        version: 2
        secauth: off
        threads: 0
        interface {
                ringnumber: 0
                bindnetaddr: 172.16.100.0
                mcastaddr: 226.94.1.1
                mcastport: 5405
                ttl: 1
        }
}
service {
        ver: 0
        name: pacemaker
}
aisexec {
        user: root
        group: root
}

logging {
        fileline: off
        to_stderr: no
        to_logfile: yes
        to_syslog: yes
        logfile: /var/log/cluster/corosync.log
        debug: off
        timestamp: on
        logger_subsys {
                subsys: AMF
                debug: off
        }
}

amf {
        mode: disabled
}

Generate corosync key file:

corosync-keygen

Install DRBD and configure global and resource files (the partition /dev/sdb1 is prepared on both hosts):

rpm -ivh drbd-8.4.3-33.el6.x86_64.rpm drbd-kmdl-2.6.32-431.el6-8.4.3-33.el6.x86_64.rpm

vim /etc/drbd.d/global_common.conf

global {
        usage-count no;
}

common {
        protocol C;
        handlers {
                # example handlers – customize as needed
        }
        startup {}
        options {}
        disk {
                on-io-error detach;
        }
        net {
                cram-hmac-alg "sha1";
                shared-secret "mydrbdlab";
        }
        syncer {
                rate 1000M;
        }
}

Configure DRBD resource for MySQL data:

vim /etc/drbd.d/mysqldata.res
resource mysqldata {
        on www.bwei.com {
                device /dev/drbd0;
                disk /dev/sdb1;
                address 172.16.100.1:7789;
                meta-disk internal;
        }
        on www2.bwei.com {
                device /dev/drbd0;
                disk /dev/sdb1;
                address 172.16.100.5:7789;
                meta-disk internal;
        }
}

drbdadm create-md mysqldata

Start DRBD service and check status:

service drbd start

drbd-overview
# Example output:
0:mysqldata/0  Connected Secondary/Secondary Inconsistent/Inconsistent C r-----

Promote one node to primary:

# drbdadm -- --overwrite-data-of-peer primary mysqldata
# drbd-overview
0:mysqldata/0  Connected Primary/Secondary UpToDate/UpToDate C r-----

Format the DRBD device and create MySQL data directory:

mke2fs -t ext4 /dev/drbd0
mkdir /data
mount /dev/drbd0 /data
mkdir /data/mysql

Create MySQL user/group and set ownership:

groupadd -r mysql -g 360
useradd mysql -g mysql -u 360 -r
chown mysql:mysql /data/mysql/

Initialize MySQL (MariaDB) on the primary node:

tar xf mariadb-10.0.10-linux-x86_64.tar.gz -C /usr/local/
ln -sv mariadb-10.0.10-linux-x86_64 mysql
/usr/local/mysql/scripts/mysql_install_db --user=mysql --datadir=/data/mysql --basedir=/usr/local/mysql/
cp mysql/support-files/my-huge.cnf /etc/my.cnf
vim /etc/my.cnf   # set datadir = /data/mysql
cp mysql/support-files/mysql.server /etc/init.d/mysqld
chkconfig --add mysqld
chkconfig mysqld off
service mysqld start
mysql/bin/mysql   # verify connection

The secondary host must receive the same MySQL configuration files and startup script but does not repeat the initialization.

Install crmsh and define cluster resources:

rpm -ivh crmsh-1.2.6-4.el6.x86_64.rpm pssh-2.3.1-2.el6.x86_64.rpm
service corosync start
crm
# Inside crm shell:
configure property no-quorum-policy=ignore
configure property stonith-enabled=false
configure primitive mysqldrbd ocf:linbit:drbd params drbd_resource=mysqldata op monitor interval=30 role=Master timeout=30 op monitor interval=40 role=Slave timeout=30 op start timeout=240 op stop timeout=100
configure primitive mysqlfs ocf:heartbeat:Filesystem params device=/dev/drbd0 directory=/data/mysql fstype=ext4 op start timeout=60 op stop timeout=60
configure primitive mysqld lsb:mysqld
configure primitive mysqlvip ocf:heartbeat:IPaddr params ip=172.16.100.4
configure ms ms_mysqldrbd mysqldrbd meta master-max=1 master-node-max=1 clone-max=2 clone-node-max=1 notify=true
configure colocation mysqld_with_mysqlfs inf: mysqld mysqlfs
configure order mysqlfs_after_ms_mysqldrbd inf: ms_mysqldrbd:promote mysqlfs:start
configure order mysqld_after_mysqlfs inf: mysqlfs mysqld
commit
crm status

The services have been successfully configured!