基于DRBD+corosync做Mysql的高可用集群服务

如果查看此篇博客文章请先看本人的上篇基于DRBD的数据同步，有什么意见希望给我留言，谢谢你的阅读！

八、安装MYSQL

1、在节点1上安装MYSQL所以要把节点1做为主节点

#把节点2降级为次节点

[root@node2 ~]# drbdadm secondary web

[root@node2 ~]# umount /dev/drbd0

[root@node2 drbd.d]# drbd-overview

  0:web  Connected Secondary/Secondary UpToDate/UpToDate C r----

#把节点1升级为主节点并挂载

[root@node1 ~]# drbdadm primary web

[root@node1 ~]# mount /dev/drbd0 /web

[root@node1 ~]# drbd-overview 

  0:web  Connected Primary/Secondary UpToDate/UpToDate C r---- /web ext3 1.9G 36M 1.8G 2%

#删除此前为了验证放在目录里的RPM包

[root@node1 ~]# cd /web

[root@node1 web]# rm -rf *.rpm

2、创建MYSQL的数据目录

#创建MYSQL的数据目录

[root@node1 web]# mkdir -pv mydata/data

mkdir: created directory `mydata'

mkdir: created directory `mydata/data'

创建mysql组和用户

#在节点1上创建

[root@node1 web]# groupadd -r -g 306 mysql

[root@node1 web]# useradd -g mysql -r -s /sbin/nologin -M -u 306 mysql

[root@node1 web]# id mysql

uid=306(mysql) gid=306(mysql) groups=306(mysql)

#在节点2上创建

[root@node2 ~]# groupadd -r -g 306 mysql

[root@node2 ~]# useradd -g mysql -r -s /sbin/nologin -M -u 306 mysql

[root@node2 ~]# id mysql

uid=306(mysql) gid=306(mysql) groups=306(mysql)

说明：在节点1和节点2上的mysql组和用户必须完全一样

#把MYSQL的数据目录的属组属主改为mysql

[root@node1 web]# chown -R mysql:mysql /web/mydata

[root@node1 web]# ll /web

total 20

drwx------ 2 root  root  16384 Dec 19 05:56 lost+found

drwxr-xr-x 3 mysql mysql  4096 Dec 20 01:46 mydata

#验证在节点2上是否和节点1上一致

#先将节点1降级为次节点

[root@node1 ~]# umount /dev/drbd0

[root@node1 ~]# mount

[root@node1 ~]# drbdadm secondary web

[root@node1 ~]# drbd-overview 

  0:web  Connected Secondary/Secondary UpToDate/UpToDate C r---- 

#将节点升级为主节点并查看目录

[root@node2 ~]# drbdadm primary web

[root@node2 ~]# mount /dev/drbd0 /web

[root@node2 ~]# mount

[root@node2 ~]# ll /web

total 20

drwx------ 2 root  root  16384 Dec 19 05:56 lost+found

drwxr-xr-x 3 mysql mysql  4096 Dec 20 01:46 mydata

说明：节点2上的MYSQL数据目录属主和属组一致 说明一切正常

3、准备MYSQL安装包和安装

#准备二进制安装包

mysql-5.5.19-linux2.6-i686.tar.gz

#在主节点上解压并安装

[root@node2 ~]# tar vxf mysql-5.5.19-linux2.6-i686.tar.gz -C /usr/local

[root@node2 ~]# cd /usr/local

[root@node2 local]# ln -sv mysql-5.5.19-linux2.6-i686/ mysql

[root@node2 local]# cd mysql

[root@node2 mysql]# chown -R mysql:mysql .

[root@node2 mysql]# ll

#初始化数据库

[root@node2 mysql]# scripts/mysql_install_db --user=mysql --datadir=/web/mydata/data

[root@node2 mysql]# ls /web/mydata/data

mysql  performance_schema  test

#修改mysql目录的属主为root

[root@node2 mysql]# chown -R root /usr/local/mysql/

[root@node2 mysql]# ll

#配置主配置文件

[root@node2 mysql]# cd support-files/

[root@node2 support-files]# cp my-large.cnf /etc/my.cnf

[root@node2 support-files]# vim /etc/my.cnf

#修改以下几项

thread_concurrency = 2

datadir = /web/mydata/data #此行是添加的

#配置使用命令启动MYSQL

[root@node2 support-files]# cp mysql.server /etc/init.d/mysqld

[root@node2 support-files]# chkconfig --add mysqld

[root@node2 support-files]# service mysqld start

Starting MySQL.........                                    [  OK  ]

#查看MYSQL的监听端口是否开启

[root@node2 support-files]# netstat -tnlp

#添加MYSQL客户端的PATH变量

[root@node2 support-files]# vim /etc/profile

#添加此行

PATH=$PATH:/usr/local/mysql/bin

#重读此配置文件

[root@node2 support-files]# . /etc/profile

#连接MYSQL服务器

[root@node2 support-files]# mysql

#创建mydb数据库备以后验证使用

mysql> create database mydb;

Query OK, 1 row affected (0.00 sec)

mysql> show databases;

+--------------------+

| Database           |

+--------------------+

| information_schema |

| mydb               |

| mysql              |

| performance_schema |

| test               |

+--------------------+

5 rows in set (0.00 sec)

说明：如果连接创建成功说明mysql安装成功

#此时关闭mysql服务

[root@node2 support-files]# cd

[root@node2 ~]# service mysqld stop

Shutting down MySQL..                                      [  OK  ]

4、在次节点我们如何安装mysql呢?

首先把mysql安装包拷到次节点上

[root@node2 ~]scp mysql-5.5.19-linux2.6-i686.tar.gz node1:/root

在节点1上安装配置mysql

[root@node1 ~]# tar vxf mysql-5.5.19-linux2.6-i686.tar.gz -C /usr/local

[root@node1 ~]# cd /usr/local

[root@node1 local]# ln -sv mysql-5.5.19-linux2.6-i686/ mysql

[root@node1 mysql]# chown mysql:mysql /usr/local/mysql/*

[root@node1 mysql]# chown root /usr/local/mysql/*

[root@node1 mysql]# ll

说明：此时目录下的文件属主属组和节点2的一致

#把节点2修改的配置文件拷贝到节点1上

[root@node2 ~]# scp /etc/my.cnf node1:/etc

[root@node2 ~]# scp /etc/init.d/mysqld node1:/etc/init.d/

[root@node2 ~]# scp /etc/profile node1:/etc

#重读此配置文件

[root@node1 ~]# . /etc/profile

此时节点1还不能启动mysql，因此我们要将节点1变为主节点

#节点1操作

[root@node2 ~]# umount /dev/drbd0

[root@node2 ~]# drbdadm secondary web

[root@node2 ~]# drbd-overview 

  0:web  Connected Secondary/Secondary UpToDate/UpToDate C r----

#节点2操作

[root@node1 ~]# drbdadm primary web

[root@node1 ~]# mount /dev/drbd0 /web

[root@node1 ~]# drbd-overview 

  0:web  Connected Primary/Secondary UpToDate/UpToDate C r---- /web ext3 1.9G 65M 1.7G 4%

#查看节点1上的数据目录是否有文件

[root@node1 ~]# ll /web/mydata/data/

total 28752

-rw-rw---- 1 mysql mysql 18874368 Dec 20 03:01 ibdata1

-rw-rw---- 1 mysql mysql  5242880 Dec 20 03:01 ib_logfile0

-rw-rw---- 1 mysql mysql  5242880 Dec 20 02:34 ib_logfile1

drwx------ 2 mysql mysql     4096 Dec 20 02:53 mydb

drwx------ 2 mysql root      4096 Dec 20 02:22 mysql

-rw-rw---- 1 mysql mysql      126 Dec 20 02:42 mysql-bin.000001

-rw-rw---- 1 mysql mysql      209 Dec 20 03:01 mysql-bin.000002

-rw-rw---- 1 mysql mysql       38 Dec 20 02:49 mysql-bin.index

-rw-rw---- 1 mysql root      3472 Dec 20 03:01 node2.wang.com.err

drwx------ 2 mysql mysql     4096 Dec 20 02:22 performance_schema

drwx------ 2 mysql root      4096 Dec 20 02:22 test

#启动mysql服务

[root@node1 ~]# service mysqld start

Starting MySQL.......                                      [  OK  ]

[root@node1 ~]# mysql

mysql> show databases;

+--------------------+

| Database           |

+--------------------+

| information_schema |

| mydb               |

| mysql              |

| performance_schema |

| test               |

+--------------------+

5 rows in set (0.01 sec)

说明：查看节点1上是否有刚我们在节点2上创建的mydb数据库

此时mysql的高可能集群都以完成，但都是我们手动操作的，那么如何让它自动切换呢

要把以下三个资源作为高可用资源

mysqld

drbd的web资源

/dev/drbd0 /web 的挂载

九、安装配置corosync

准备以下安装包

cluster-glue-1.0.6-1.6.el5.i386.rpm

cluster-glue-libs-1.0.6-1.6.el5.i386.rpm

corosync-1.2.7-1.1.el5.i386.rpm

corosynclib-1.2.7-1.1.el5.i386.rpm

heartbeat-3.0.3-2.3.el5.i386.rpm

heartbeat-libs-3.0.3-2.3.el5.i386.rpm

libesmtp-1.0.4-5.el5.i386.rpm

mysql-5.5.19-linux2.6-i686.tar.gz

openais-1.1.3-1.6.el5.i386.rpm

openaislib-1.1.3-1.6.el5.i386.rpm

pacemaker-1.0.11-1.2.el5.i386.rpm

pacemaker-libs-1.0.11-1.2.el5.i386.rpm

perl-TimeDate-1.16-5.el5.noarch.rpm

resource-agents-1.0.4-1.1.el5.i386.rpm

1、#安装

[root@node1 ~]# yum --nogpgcheck -y localinstall *.rpm

[root@node1 ~]# scp *.rpm node2:/root

[root@node2 ~]# yum --nogpgcheck -y localinstall *.rpm

2、配置文件

[root@node1 corosync]# cp corosync.conf.example corosync.conf

[root@node1 corosync]# vim corosync.conf

totem {

        version: 2

        secauth: on

        threads: 0

        interface {

                ringnumber: 0

                bindnetaddr: 172.16.1.0

                mcastaddr: 226.94.1.1

                mcastport: 5405

        }

}

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

}

amf {

        mode: disabled

}

service {

        ver: 0

        name: pacemaker

}

aisexec {

        user: root

        group:root

}

#创建corosync的认证密钥

[root@node1 corosync]# corosync-keygen

#将配置文件拷贝到节点2上

[root@node1 corosync]# scp -p authkey corosync.conf node2:/etc/corosync

#创建日志存放目录

[root@node1 corosync]# mkdir /var/log/cluster

[root@node1 corosync]# ssh node2 'mkdir /var/log/cluster'

3、#启动corosync服务

[root@node1 ~]# service corosync start

Starting Corosync Cluster Engine (corosync):               [  OK  ]

#查看日志信息

[root@node1 ~]# tail /var/log/cluster/corosync.log

#启动节点2的corosync服务

[root@node1 ~]# ssh node2 'service corosync start'

Starting Corosync Cluster Engine (corosync): [  OK  ]

#查看服务启动状态

[root@node1 ~]# crm status

============

Last updated: Tue Dec 20 04:00:18 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

0 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

4、配置资源

#查看当前配置信息

[root@node1 ~]# crm

crm(live)# configure show

INFO: building help index

node node1.wang.com

node node2.wang.com

property $id="cib-bootstrap-options" \

dc-version="1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87" \

cluster-infrastructure="openais" \

expected-quorum-votes="2"

crm(live)# configure

crm(live)configure# property stonith-enabled=false   #禁用stonith设备

crm(live)configure# verify     #验证是否出错

crm(live)configure# property no-quorum-policy=ignore #忽略法定票数

crm(live)configure# verify 

crm(live)configure# rsc_defaults resource-stickiness=100 #配置资源粘性值

crm(live)configure# verify

crm(live)configure# commit     #提交操作

crm(live)configure# show     #查看操作是否生效

node node1.wang.com

node node2.wang.com

property $id="cib-bootstrap-options" \

dc-version="1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87" \

cluster-infrastructure="openais" \

expected-quorum-votes="2" \

stonith-enabled="false" \

no-quorum-policy="ignore"

rsc_defaults $id="rsc-options" \

resource-stickiness="100"

配置drbd集群资源

#先关闭所有要添加高可用集群的资源，并开机不能自动启动

[root@node1 ~]# service mysqld stop

Shutting down MySQL....                                    [  OK  ]

[root@node1 ~]# chkconfig mysqld off

[root@node1 ~]# umount /dev/drbd0

[root@node1 ~]# drbdadm secondary web

[root@node1 ~]# service drbd stop

Stopping all DRBD resources: .

[root@node1 ~]# chkconfig drbd off

说明：在节点2上做同样的操作

#开始定义资源

[root@node1 ~]# crm

crm(live)# configure

crm(live)configure# primitive webdrbd ocf:linbit:drbd params drbd_resource=web

crm(live)configure# master ms_webdrbd webdrbd meta master-max=1 master-nodes-max=1 clone-max=2 clone-nodes-max=1 notify=true

crm(live)configure# verify

crm(live)configure# commit

#查看定义的资源

crm(live)configure# show

node node1.wang.com

node node2.wang.com

primitive webdrbd ocf:linbit:drbd \

params drbd_resource="web"

ms ms_webdrbd webdrbd \

meta master-max="1" master-nodes-max="1" clone-max="2" clone-nodes-max="1" notify="true"

property $id="cib-bootstrap-options" \

dc-version="1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87" \

cluster-infrastructure="openais" \

expected-quorum-votes="2" \

stonith-enabled="false" \

no-quorum-policy="ignore"

rsc_defaults $id="rsc-options" \

resource-stickiness="100"

#查看状态

crm(live)# status 

============

Last updated: Tue Dec 20 04:31:58 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

1 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node1.wang.com ]

     Slaves: [ node2.wang.com ]

说明：此时节点1为主节点

#切换主节点

[root@node1 ~]# crm node standby

[root@node1 ~]# crm status

============

Last updated: Tue Dec 20 04:48:11 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

1 Resources configured.

============

Node node1.wang.com: standby

Online: [ node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node2.wang.com ] #节点2为主节点

     Stopped: [ webdrbd:0 ] #节点1是standby因此只有主节点

#在节点2上查看主次状态

[root@node2 ~]# drbd-overview 

  0:web  WFConnection Primary/Unknown UpToDate/Outdated C r----

#让节点1上线

[root@node1 ~]# crm node online

[root@node1 ~]# crm status

============

Last updated: Tue Dec 20 04:51:45 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

1 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node2.wang.com ]

     Slaves: [ node1.wang.com ]

在主节点上定义资源，此时主节点在节点2上

[root@node2 ~]# crm

crm(live)# configure

crm(live)configure# primitive webfs ocf:heartbeat:Filesystem params device="/dev/drbd0" directory="/web" fstype="ext3"

定义资源约束

#定义位置约束

crm(live)configure# colocation webfs_on_ms_webdrbd inf: webfs ms_webdrbd:Master

#定义顺序约束

crm(live)configure# order webfs_after_ms_webdrbd inf: ms_webdrbd:promote webfs:start

#定义排列约束

crm(live)configure# verify

crm(live)configure# commit

crm(live)configure# show

node node1.wang.com \

attributes standby="off"

node node2.wang.com

primitive webdrbd ocf:linbit:drbd \

params drbd_resource="web"

primitive webfs ocf:heartbeat:Filesystem \

params device="/dev/drbd0" directory="/web" fstype="ext3"

ms ms_webdrbd webdrbd \

meta master-max="1" master-nodes-max="1" clone-max="2" clone-nodes-max="1" notify="true"

location drbd-fence-by-handler-ms_webdrbd ms_webdrbd \

rule $id="drbd-fence-by-handler-rule-ms_webdrbd" $role="Master" -inf: #uname ne node2.wang.com

colocation webfs_on_ms_webdrbd inf: webfs ms_webdrbd:Master

order webfs_after_ms_webdrbd inf: ms_webdrbd:promote webfs:start

property $id="cib-bootstrap-options" \

dc-version="1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87" \

cluster-infrastructure="openais" \

expected-quorum-votes="2" \

stonith-enabled="false" \

no-quorum-policy="ignore"

rsc_defaults $id="rsc-options" \

resource-stickiness="100"

#查看资源运行状态

crm(live)# status

============

Last updated: Tue Dec 20 05:08:12 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

2 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node2.wang.com ]

     Slaves: [ node1.wang.com ]

 webfs (ocf::heartbeat:Filesystem): Started node2.wang.com

#定义mysql 集群资源

crm(live)# configure

crm(live)configure# primitive webmysqld lsb:mysqld

crm(live)configure# order mysqld_after_webfs inf: webfs:start webmysqld：start

crm(live)configure# colocation mysqld_on_webfs inf: webmysqld webfs

crm(live)configure# verify

crm(live)configure# commit

crm(live)# status 

============

Last updated: Tue Dec 20 05:23:48 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

3 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node2.wang.com ]

     Slaves: [ node1.wang.com ]

 webfs (ocf::heartbeat:Filesystem): Started node2.wang.com

crm(live)# status 

============

Last updated: Tue Dec 20 05:24:17 2011

Stack: openais

Current DC: node1.wang.com - partition with quorum

Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87

2 Nodes configured, 2 expected votes

3 Resources configured.

============

Online: [ node1.wang.com node2.wang.com ]

 Master/Slave Set: ms_webdrbd

     Masters: [ node2.wang.com ]

     Slaves: [ node1.wang.com ]

 webfs (ocf::heartbeat:Filesystem): Started node2.wang.com

 webmysqld (lsb:mysqld): Started node2.wang.com

说明：此时mysql资源在主节点上启动起来

此时切换节点查看资源状态

mysql基于brbd+corosync的高可用集群就完成了