corosync+pacemaker+mysql+drbd 实现mysql的高可用

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

一、了解各组件是什么以及之间的关系

二、安装高可用集群的提前

三、安装corosync+pacemaker

四、编译安装mysql

五、安装drbd

六、mysql与drbd实现mysql数据的镜像

七、利用crmsh配置mysql的高可用

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

环境:

OS:Centos 6.x(redhat 6.x)

kernel:2.6.32-358.el6.x86_64

yum源:

1
2
3
4
5
6
7
8
9
10
11
[centos]
name=sohu-centos
baseurl=http://mirrors.sohu.com/centos/$releasever/os/$basearch
gpgcheck=1
enable=0
gpgkey=http://mirrors.sohu.com/centos/RPM-GPG-KEY-CentOS-6
[epel]
name=sohu-epel
baseurl=http://mirrors.sohu.com/fedora-epel/$releasever/$basearch/
enable=1
gpgcheck=0


拓扑图:

205713233.png

部分软件以附件的形式上传,mysql的源码包软件,网上很容易找


望各位博友带着下面的疑问去实现corosync+pacemaker+mysql+drbd mysql的高可用?

1、corosync是什么?pacemaker是什么?corosync与pacemaker的关系?

2、 mysql与drbd之间的连接关系?

3、 corosync、pacemaker、mysql、drbd之间的关系?

4、提到高可用集群,广大博友们就会想到,高可用集群中资源之间是如何建立关系?各个节点会不会抢占资源使其出现脑裂(split-brain)?

出现脑裂可以有fence设备自行解决


一、了解各组件是什么以及之间的关系

corosynccorosync的由来是源于一个Openais的项目,是Openais的一个子 项目,可以实现HA心跳信息传输的功能,是众多实现HA集群软件中之一,heartbeat与corosync是流行的Messaging Layer (集群信息层)工具。而corosync是一个新兴的软件,相比Heartbeat这款很老很成熟的软件,corosync与Heartbeat各有优势,博主就不在这里比较之间的优势了,corosync相对于Heartbeat只能说现在比较流行。

pacemaker:是众多集群资源管理器(Cluster Resource Manager)CRM中的一个,其主要功能是管理来着集群信息层发送来的信息。Pacemaker是集群的核心,它管理集群和集群信息. 集群信息更新通过Corosync通知到各个节点.

常见的CRM有:

heartbeat v1-->haresources

hearbeat v2--->crm

hearbeat v3---->pacemaker

RHCS(cman)----->rgmanager

如下图corosync与pacemaker之间的关系:

161511846.png


mysql:一个开源的关系型数据库

drbd:DRBD:(distributed replication block device)即分布式复制块设备。它的工作原理是:在A主机上有对指定磁盘设备写请求时,数据发送给A主机的kernel,然后通kernel中的一个模块,把相同的数据传送给B主机的kernel中一份,然后B主机再写入自己指定的磁盘设备,从而实现两主机数据的同步,也就实现了写操作高可用。类似于raid1一样,实现数据的镜像,DRBD一般是一主一从,并且所有的读写操作,挂载只能在主节点服务器上进行,,但是主从DRBD服务器之间是可以进行调换的。

各个组件之间的关系:其实mysql与drbd根本没有半毛钱的关系,而drbd与mysql相结合就有很重要的作用了,因为drbd实现数据的镜像,当drbd的主节点挂了之后,drbd的辅助节点还可以提供服务,但是主节点不会主动的切换到辅助的节点上面去,于是乎,高可用集群就派上用场了,因为资源定义为高可用的资源,主节点出现故障之后,高可用集群可以自动的切换到辅助节点上去,实现故障转移继续提供服务。

二、安装高可用集群的提前准备

1)、hosts文件

1
2
3
4
5
6
7
8
9
10
11
12
     #把主机名改成jie2.com
[root@jie2 ~]# sed -i s/`grep HOSTNAME /etc/sysconfig/network |awk -F '=' '{print $2}'`/jie2.com/g /etc/sysconfig/network
     #把主机名改成jie3.com
[root@jie3 ~]# sed -i s/`grep HOSTNAME /etc/sysconfig/network |awk -F '=' '{print $2}'`/jie3.com/g /etc/sysconfig/network
[root@jie2 ~]# cat >>/etc/hosts << EOF
>172.16.22.2 jie2.com  jie2
>172.16.22.3 jie3.com  jie3
>EOF
[root@jie3 ~]# cat >>/etc/hosts << EOF
>172.16.22.2 jie2.com  jie2
>172.16.22.3 jie3.com  jie3
>EOF

2)、ssh互信

1
2
3
4
[root@jie2 ~]# ssh-keygen -t rsa -P ''
[root@jie2 ~]# ssh-copy-id -i .ssh/id_rsa.pub jie3
[root@jie3 ~]# ssh-keygen -t rsa -P ''
[root@jie3 ~]# ssh-copy-id -i .ssh/id_rsa.pub jie2

3)、关闭NetworkManger

1
2
3
4
5
6
[root@jie2 ~]# chkconfig --del NetworkManager
[root@jie2 ~]# chkconfig NetworkManager off
[root@jie2 ~]# service NetworkManager stop
[root@jie3 ~]# chkconfig --del NetworkManager
[root@jie3 ~]# chkconfig NetworkManager off
[root@jie3 ~]# service NetworkManager stop


4)、时间同步(博主用的是自己的ntp时间服务器)

1
2
[root@jie2 ~]#ntpdate 172.16.0.1
[root@jie3 ~]#ntpdate 172.16.0.1


三、安装corosync+pacemaker

1)、安装corosync+pacemaker

1
2
3
4
5
6
    #节点jie2.com的操作
[root@jie2 ~]# yum -y install corosync pacemaker
[root@jie2 ~]# yum -y --nogpgcheck install crmsh-1.2.6-4.el6.x86_64.rpm pssh-2.3.1-2.el6.x86_64.rpm   #提供crmsh命令接口的软件
    #节点jie3.com的操作
[root@jie3 ~]# yum -y install corosync pacemaker
[root@jie3 ~]# yum -y --nogpgcheck install crmsh-1.2.6-4.el6.x86_64.rpm pssh-2.3.1-2.el6.x86_64.rpm

2)、修改配置文件和生成认证文件

配置文件

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
      #节点jie2.com的操作
[root@jie2 ~]# cd /etc/corosync/
[root@jie2 corosync]# mv corosync.conf.example corosync.conf
[root@jie2 corosync]# vim corosync.conf
# Please read the corosync.conf.5 manual page
compatibility: whitetank
totem {                           #心跳信息传递层
    version: 2                    #版本
    secauth: on                   #认证信息  一般on
    threads: 0                    #线程
    interface {                   #定义心跳信息传递的接口
        ringnumber: 0
        bindnetaddr: 172.16.0.0   #绑定的网络地址,写网络地址
        mcastaddr: 226.94.1.1     #多播地址
        mcastport: 5405           #多播的端口
        ttl: 1                    #生存周期
    }
}
logging {                         #日志
    fileline: off
    to_stderr: no                 #是否输出在屏幕上
    to_logfile: yes               #定义自己的日志
    to_syslog: no                 #是否由syslog记录日志
    logfile: /var/log/cluster/corosync.log  #日志文件的存放路径
    debug: off
    timestamp: on                 #时间戳是否关闭
    logger_subsys {
        subsys: AMF
        debug: off
    }
}
amf {
    mode: disabled
}
service {
     ver: 0
     name: pacemaker   #pacemaker作为corosync的插件进行工作
}
aisexec {
     user: root
     group: root
}
[root@jie2 corosync]#  scp corosync.conf jie3:/etc/corosync/
##把节点jie2.com的配置文件copy到jie3.com中

认证文件

1
2
3
4
5
6
7
8
     #节点jie2.com的操作
[root@jie2 corosync]# corosync-keygen
Corosync Cluster Engine Authentication key generator.
Gathering 1024 bits for key from /dev/random.
Press keys on your keyboard to generate entropy (bits = 152).
     #遇到这个情况,表示电脑的随机数不够,各位朋友可以不停的随便敲键盘,或者安装软件也可以生成随机数
[root@jie2 corosync]# scp authkey jie3:/etc/corosync/
     #把认证文件也复制到jie3.com主机上

3)、开启服务和查看集群中的节点信息

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
         #节点jie2.com的操作
[root@jie2 ~]# service corosync start
Starting Corosync Cluster Engine (corosync):               [  OK  ]
[root@jie2 ~]# crm status
Last updated: Thu Aug  8 14:43:13 2013
Last change: Sun Sep  1 16:41:18 2013 via crm_attribute on jie3.com
Stack: classic openais (with plugin)
Current DC: jie3.com - partition with quorum
Version: 1.1.8-7.el6-394e906
2 Nodes configured, 2 expected votes
Online: [ jie2.com jie3.com ]
         #节点jie3.com的操作
[root@jie3 ~]# service corosync start
Starting Corosync Cluster Engine (corosync):               [  OK  ]
[root@jie3 ~]# crm status
Last updated: Thu Aug  8 14:43:13 2013
Last change: Sun Sep  1 16:41:18 2013 via crm_attribute on jie3.com
Stack: classic openais (with plugin)
Current DC: jie3.com - partition with quorum
Version: 1.1.8-7.el6-394e906
2 Nodes configured, 2 expected votes
Online: [ jie2.com jie3.com ]


四、编译安装mysql(两个节点的操作过程都是一样)


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
     #节点jie2.com的操作
#1)、解压编译安装
[root@jie2 ~]# tar xf mysql-5.5.33.tar.gz
[root@jie2 ~]# yum -y groupinstall "Development tools" "Server Platform Development"
[root@jie2 ~]# cd mysql-5.5.33
[root@jie2 mysql-5.5.33]# yum -y install cmake
[root@jie2 mysql-5.5.33]# cmake . -DCMAKE_INSTALL_PREFIX=/usr/local/mysql \
-DMYSQL_DATADIR=/mydata/data  -DSYSCONFDIR=/etc \
-DWITH_INNOBASE_STORAGE_ENGINE=1 -DWITH_ARCHIVE_STORAGE_ENGINE=1 \
-DWITH_BLACKHOLE_STORAGE_ENGINE=1 -DWITH_READLINE=1 -DWITH_SSL=system \
-DWITH_ZLIB=system -DWITH_LIBWRAP=0 -DMYSQL_UNIX_ADDR=/tmp/mysql.sock \
-DDEFAULT_CHARSET=utf8 -DDEFAULT_COLLATION=utf8_general_ci
[root@jie2 mysql-5.5.33]# make && make install
#2)、建立配置文件和脚本文件
[root@jie2 mysql-5.5.33]# cp /usr/local/mysql/support-files/my-large.cnf /etc/my.cnf
[root@jie2 mysql-5.5.33]# cp /usr/local/mysql/support-files/mysql.server  /etc/rc.d/init.d/mysqld
[root@jie2 mysql-5.5.33]# cd /usr/local/mysql/
[root@jie2 mysql]# useradd -r -u 306 mysql
[root@jie2 mysql]# chown -R root:mysql ./*
#3)、关联系统识别的路径
[root@jie2 mysql]#echo "PATH=/usr/local/mysql/bin:$PATH" >/etc/profile.d/mysqld.sh
[root@jie2 mysql]#source /etc/profile.d/mysqld.sh
[root@jie2 mysql]#echo "/usr/local/mysql/lib" >/etc/ld.so.conf.d/mysqld.conf
[root@jie2 mysql]#ldconfig -v | grep mysql
[root@jie2 mysql]#ln -sv /usr/local/mysql/include/ /usr/local/mysqld


先别初始化数据库,安装drbd把drbd挂载到目录下,然后初始化数据库把数据库的数据存放到drbd挂载的目录。

五、安装drbd

安装rpm包的drbd软件必须保证找相同内核版本的drbd-kmdl软件

1)、先划分一个分区,此分区做成drbd镜像(RHEL 6.x的重新格式化一个新的分区之后要重启系统)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
       #节点jie2.com的操作
[root@jie2 ~]# fdisk /dev/sda
Command (m for help): n
Command action
   e   extended
   p   primary partition (1-4)
p
Partition number (1-4): 3
First cylinder (7859-15665, default 7859):
Using default value 7859
Last cylinder, +cylinders or +size{K,M,G} (7859-15665, default 15665): +5G
Command (m for help): w
         #节点jie3.com的操作
[root@jie3 ~]# fdisk /dev/sda
Command (m for help): n
Command action
   e   extended
   p   primary partition (1-4)
p
Partition number (1-4): 3
First cylinder (7859-15665, default 7859):
Using default value 7859
Last cylinder, +cylinders or +size{K,M,G} (7859-15665, default 15665): +5G
Command (m for help): w


2)、安装drbd和修改配置文件

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
#1)、安装drbd
     #节点jie2.com的操作
[root@jie2 ~]# rpm -ivh drbd-kmdl-2.6.32-358.el6-8.4.3-33.el6.x86_64.rpm
warning: drbd-kmdl-2.6.32-358.el6-8.4.3-33.el6.x86_64.rpm: Header V4 DSA/SHA1 Signature, key ID 66534c2b: NOKEY
Preparing...                ################################# [100%]
[root@jie2 ~]# rpm -ivh drbd-8.4.3-33.el6.x86_64.rpm
warning: drbd-8.4.3-33.el6.x86_64.rpm: Header V4 DSA/SHA1 Signature, key ID 66534c2b: NOKEY
Preparing...                ################################## [100%]
     #节点jie3.com的操作
[root@jie3 ~]# rpm -ivh drbd-kmdl-2.6.32-358.el6-8.4.3-33.el6.x86_64.rpm
warning: drbd-kmdl-2.6.32-358.el6-8.4.3-33.el6.x86_64.rpm: Header V4 DSA/SHA1 Signature, key ID 66534c2b: NOKEY
Preparing...                ################################# [100%]
[root@jie3 ~]# rpm -ivh drbd-8.4.3-33.el6.x86_64.rpm
warning: drbd-8.4.3-33.el6.x86_64.rpm: Header V4 DSA/SHA1 Signature, key ID 66534c2b: NOKEY
Preparing...                ################################## [100%]
#2)、修改drbd的配置文件
      #节点jie2.com的操作
[root@jie2 ~]# cd /etc/drbd.d/
[root@jie2 drbd.d]# cat global_common.conf  #全局配置文件
global {
        usage-count no;
        # minor-count dialog-refresh disable-ip-verification
}
common {
        protocol C;
        handlers {
                pri-on-incon-degr "/usr/lib/drbd/notify-pri-on-incon-degr.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /proc/sysrq-trigger ; reboot -f";
                pri-lost-after-sb "/usr/lib/drbd/notify-pri-lost-after-sb.sh; /usr/lib/drbd/notify-emergency-reboot.sh; echo b > /proc/sysrq-trigger ; reboot -f";
                local-io-error "/usr/lib/drbd/notify-io-error.sh; /usr/lib/drbd/notify-emergency-shutdown.sh; echo o > /proc/sysrq-trigger ; halt -f";
                # fence-peer "/usr/lib/drbd/crm-fence-peer.sh";
                # split-brain "/usr/lib/drbd/notify-split-brain.sh root";
                # out-of-sync "/usr/lib/drbd/notify-out-of-sync.sh root";
                # before-resync-target "/usr/lib/drbd/snapshot-resync-target-lvm.sh -p 15 -- -c 16k";
                # after-resync-target /usr/lib/drbd/unsnapshot-resync-target-lvm.sh;
        }
        startup {
                #wfc-timeout 120;
                #degr-wfc-timeout 120;
        }
        disk {
                on-io-error detach;
                #fencing resource-only;
        }
        net {
                cram-hmac-alg "sha1";
                shared-secret "mydrbdlab";
        }
        syncer {
                rate 1000M;
        }
}
[root@jie2 drbd.d]# cat mydata.res #资源配置文件
resource mydata {
  on jie2.com {
    device    /dev/drbd0;
    disk      /dev/sda3;
    address   172.16.22.2:7789;
    meta-disk internal;
  }
  on jie3.com {
    device    /dev/drbd0;
    disk      /dev/sda3;
    address   172.16.22.3:7789;
    meta-disk internal;
  }
}
    #把配置文件copy到节点jie3.com上面
[root@jie2 drbd.d]# scp global_common.conf mydata.res jie3:/etc/drbd.d/


3)、初始化drbd的资源并启动

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
     #节点jie2.com的操作
#创建drbd的资源
[root@jie2 ~]# drbdadm create-md mydata
Writing meta data...
initializing activity log
NOT initializing bitmap
lk_bdev_save(/var/lib/drbd/drbd-minor-0.lkbd) failed: No such file or directory
New drbd meta data block successfully created.  #提示已经创建成功
lk_bdev_save(/var/lib/drbd/drbd-minor-0.lkbd) failed: No such file or directory
#启动服务
[root@jie2 ~]# service drbd start
Starting DRBD resources: [
     create res: mydata
   prepare disk: mydata
    adjust disk: mydata
     adjust net: mydata
]
..........                [ok]
     #节点jie3.com的操作
#创建drbd的资源
[root@jie3 ~]# drbdadm create-md mydata
Writing meta data...
initializing activity log
NOT initializing bitmap
lk_bdev_save(/var/lib/drbd/drbd-minor-0.lkbd) failed: No such file or directory
New drbd meta data block successfully created.  #提示已经创建成功
lk_bdev_save(/var/lib/drbd/drbd-minor-0.lkbd) failed: No such file or directory
#启动服务
[root@jie3 ~]# service drbd start
Starting DRBD resources: [
     create res: mydata
   prepare disk: mydata
    adjust disk: mydata
     adjust net: mydata
]
..........                [ok]

4)、设置一个主节点,然后同步drbd的数据(此步骤只需在一个节点上操作)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
   #设置jie2.com为drbd的主节点
[root@jie2 ~]# drbdadm primary --force mydata
[root@jie2 ~]# cat /proc/drbd     #查看同步进度
version: 8.4.3 (api:1/proto:86-101)
GIT-hash: 89a294209144b68adb3ee85a73221f964d3ee515 build by gardner@, 2013-05-27 04:30:21
 0: cs:SyncSource ro:Primary/Secondary ds:UpToDate/Inconsistent C r---n-
    ns:1897624 nr:0 dw:0 dr:1901216 al:0 bm:115 lo:0 pe:3 ua:3 ap:0 ep:1 wo:f oos:207988
    [=================>..] sync'ed: 90.3% (207988/2103412)K
    finish: 0:00:07 speed: 26,792 (27,076) K/sec
[root@jie2 ~]# watch -n1 'cat /proc/drbd'  此命令可以动态的查看
[root@jie2 ~]# cat /proc/drbd
version: 8.4.3 (api:1/proto:86-101)
GIT-hash: 89a294209144b68adb3ee85a73221f964d3ee515 build by gardner@, 2013-05-27 04:30:21
 0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r-----
    ns:120 nr:354 dw:435 dr:5805 al:6 bm:9 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:0    #当两边都为UpToDate时,表示两边已经同步

5)、格式化drdb分区(此步骤在主节点上操作)

1
[root@jie2 ~]# mke2fs -t ext4 /dev/drbd0

六、mysql与drbd实现mysql数据的镜像

1)、在drbd的主节点上,挂载drbd的分区,然后初始化数据库

1
2
3
4
5
6
7
8
9
10
[root@jie2 ~]# mkdir /mydata  #创建用于挂载drbd的目录
[root@jie2 ~]# mount /dev/drbd0 /mydata/
[root@jie2 ~]# mkdir /mydata/data
[root@jie2 ~]#chown -R mysql.mysql /mydata  #把文件的属主和属组改成mysql
[root@jie2 ~]#vim /etc/my.cnf   #修改mysql的配置文件
        datadir = /mydata/data
        innodb_file_per_table =1
[root@jie2 ~]#/usr/local/mysql/scripts/mysql_install_db --user=mysql --datadir=/mydata/data/  --basedir=/usr/local/mysql   #初始化数据库
[root@jie2 ~]# service mysqld start
Starting MySQL .......                                [  OK  ]


2)、验证drbd是否镜像

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
      #节点jie2.com的操作
#1)、先在drbd的主节点上面创建一个数据库
[root@jie2 ~]# mysql
mysql> show databases;
+--------------------+
| Database           |
+--------------------+
| information_schema |
| mysql              |
| performance_schema |
test               |
+--------------------+
4 rows in set (0.00 sec)
mysql> create database jie2;
Query OK, 1 row affected (0.01 sec)
mysql> show databases;
+--------------------+
| Database           |
+--------------------+
| information_schema |
| jie2               |
| mysql              |
| performance_schema |
test               |
+--------------------+
5 rows in set (0.00 sec)
mysql>\q
#2)、停掉mysql服务,卸载drbd挂载的目录
[root@jie2 ~]# service mysqld stop
[root@jie2 ~]# umount /dev/drbd0 #卸载drbd的挂载点
[root@jie2 ~]# drbdadm  secondary mydata  #把此节点改为drbd的备用节点
     #节点jie3.com的操作
#3)、把jie3.com变为drbd的主节点
[root@jie3 ~]#drbdadm  primary mydata   #把此节点改为drbd的主节点
[root@jie3 ~]# mkdir /mydata
[root@jie3 ~]#chown -R mysql.mysql /mydata
[root@jie3 ~]#mount /dev/drdb0 /mydata
[root@jie3 ~]#vim /etc/my.cnf
      datadir = /mydata/data
      innodb_file_per_table = 1
[root@jie3 ~]# service mysqld start   #此节点上不用初始化数据库,直接开启服务即可
Starting MySQL .......                                [  OK  ]
[root@jie3 ~]# mysql
mysql> show databases;    #可以看见jie2数据库
+--------------------+
| Database           |
+--------------------+
| information_schema |
| jie2               |
| mysql              |
| performance_schema |
test               |
+--------------------+
5 rows in set (0.00 sec)


七、利用crmsh配置mysql的高可用

需要定义集群资源而mysql、drbd都是集群的资源,由集群管理的资源开机是一定不能够自行启动的。

1)、关闭drbd的服务和关闭mysql的服务

1
2
3
4
5
[root@jie2 ~]#service mysqld stop
[root@jie2 ~]#service drbd stop
[root@jie3 ~]#service mysqld stop
[root@jie3 ~]#umount /dev/drbd0  #之前drbd已经挂载到jie3.com节点上了
[root@jie3 ~]#service drdb stop

2)、定义集群资源

定义drbd的资源(提供drbd的资源代理RA由OCF类别中的linbit提供)

1
2
3
4
5
6
[root@jie2 ~]# crm
crm(live)# configure
crm(live)configure# property stonith-enabled=false
crm(live)configure# property no-quorum-policy=ignore
crm(live)configure# primitive mysqldrbd ocf:linbit:drbd params drbd_resource=mydata op monitor role=Master interval=10 timeout=20  op monitor  role=Slave interval=20 timeout=20 op start timeout=240 op stop timeout=100
crm(live)configure#verify   #可以检查语法

定义drbd的主从资源

1
2
crm(live)configure# ms ms_mysqldrbd mysqldrbd meta master-max=1 master-node-max=1 clone-max=2 clone-node-max=1 notify=true
crm(live)configure# verify

定义文件系统资源和约束关系

1
2
3
4
5
crm(live)configure# primitive mystore ocf:heartbeat:Filesystem params device="/dev/drbd0" directory="/mydata" fstype="ext4" op monitor interval=40 timeout=40 op start timeout=60 op stop timeout=60
crm(live)configure# verify
crm(live)configure# colocation mystore_with_ms_mysqldrbd inf: mystore ms_mysqldrbd:Master
crm(live)configure# order ms_mysqldrbd_before_mystore mandatory: ms_mysqldrbd:promote mystore:start
crm(live)configure# verify


定义vip资源、mysql服务的资源约束关系

1
2
3
4
5
6
7
8
9
10
crm(live)configure# primitive myvip ocf:heartbeat:IPaddr params ip="172.16.22.100" op monitor interval=20 timeout=20 on-fail=restart
crm(live)configure# primitive myserver lsb:mysqld op monitor interval=20 timeout=20 on-fail=restart
crm(live)configure# verify
crm(live)configure# colocation myserver_with_mystore inf: myserver mystore
crm(live)configure# order mystore_before_myserver mandatory: mystore:start myserver:start
crm(live)configure# verify
crm(live)configure# colocation myvip_with_myserver inf: myvip myserver
crm(live)configure# order myvip_before_myserver mandatory: myvip myserver
crm(live)configure# verify
crm(live)configure# commit


查看所有定义资源的信息

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
crm(live)configure# show
node jie2.com \
        attributes standby="off"
node jie3.com \
        attributes standby="off"
primitive myserver lsb:mysqld \
        op monitor interval="20" timeout="20" on-fail="restart"
primitive mysqldrbd ocf:linbit:drbd \
        params drbd_resource="mydata" \
        op monitor role="Master" interval="10" timeout="20" \
        op monitor role="Slave" interval="20" timeout="20" \
        op start timeout="240" interval="0" \
        op stop timeout="100" interval="0"
primitive mystore ocf:heartbeat:Filesystem \
        params device="/dev/drbd0" directory="/mydata" fstype="ext4" \
        op monitor interval="40" timeout="40" \
        op start timeout="60" interval="0" \
        op stop timeout="60" interval="0"
primitive myvip ocf:heartbeat:IPaddr \
        params ip="172.16.22.100" \
        op monitor interval="20" timeout="20" on-fail="restart" \
        meta target-role="Started"
ms ms_mysqldrbd mysqldrbd \
        meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1" notify="true"
colocation myserver_with_mystore inf: myserver mystore
colocation mystore_with_ms_mysqldrbd inf: mystore ms_mysqldrbd:Master
colocation myvip_with_myserver inf: myvip myserver
order ms_mysqldrbd_before_mystore inf: ms_mysqldrbd:promote mystore:start
order mystore_before_myserver inf: mystore:start myserver:start
order myvip_before_myserver inf: myvip myserver
property $id="cib-bootstrap-options" \
        dc-version="1.1.8-7.el6-394e906" \
        cluster-infrastructure="classic openais (with plugin)" \
        expected-quorum-votes="2" \
        stonith-enabled="false" \
        no-quorum-policy="ignore"

查看资源运行的状态运行在jie3.com上

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
[root@jie2 ~]# crm status
Last updated: Thu Aug  8 17:55:30 2013
Last change: Sun Sep  1 16:41:18 2013 via crm_attribute on jie3.com
Stack: classic openais (with plugin)
Current DC: jie3.com - partition with quorum
Version: 1.1.8-7.el6-394e906
2 Nodes configured, 2 expected votes
5 Resources configured.
Online: [ jie2.com jie3.com ]
 Master/Slave Set: ms_mysqldrbd [mysqldrbd]
     Masters: [ jie3.com ]
     Slaves: [ jie2.com ]
 mystore    (ocf::heartbeat:Filesystem):    Started jie3.com
 myvip  (ocf::heartbeat:IPaddr):    Started jie3.com
 myserver   (lsb:mysqld):   Started jie3.com


切换节点,看资源是否转移

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
[root@jie3 ~]# crm node standby jie3.com   #把此节点设置为备用节点
[root@jie3 ~]# crm status
Last updated: Mon Sep  2 01:45:07 2013
Last change: Mon Sep  2 01:44:59 2013 via crm_attribute on jie3.com
Stack: classic openais (with plugin)
Current DC: jie3.com - partition with quorum
Version: 1.1.8-7.el6-394e906
2 Nodes configured, 2 expected votes
5 Resources configured.
Node jie3.com: standby
Online: [ jie2.com ]
 Master/Slave Set: ms_mysqldrbd [mysqldrbd]
     Masters: [ jie2.com ]   #资源已然转到jie2.com上面
     Stopped: [ mysqldrbd:1 ]
 mystore    (ocf::heartbeat:Filesystem):    Started jie2.com
 myvip  (ocf::heartbeat:IPaddr):    Started jie2.com
 myserver   (lsb:mysqld):   Started jie2.com

由于定义了drbd的资源约束,Masters运行在那个节点,则此节点不可能成为drbd的辅助节点

1
2
3
4
5
6
7
8
[root@jie3 ~]# cat /proc/drbd
version: 8.4.3 (api:1/proto:86-101)
GIT-hash: 89a294209144b68adb3ee85a73221f964d3ee515 build by gardner@, 2013-05-27 04:30:21
 0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r-----
    ns:426 nr:354 dw:741 dr:6528 al:8 bm:9 lo:0 pe:0 ua:0 ap:0 ep:1 wo:f oos:0
[root@jie3 ~]# drbdadm secondary mydata
0: State change failed: (-12) Device is held open by someone
Command 'drbdsetup secondary 0' terminated with exit code 11


手动的停掉myvip资源还是会启动(因为定义资源是指定了on-fail=restart)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
[root@jie2 ~]# ifconfig | grep eth0
eth0      Link encap:Ethernet  HWaddr 00:0C:29:1F:74:CF
          inet addr:172.16.22.2  Bcast:172.16.255.255  Mask:255.255.0.0
          inet6 addr: fe80::20c:29ff:fe1f:74cf/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2165062 errors:0 dropped:0 overruns:0 frame:0
          TX packets:4109895 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:167895762 (160.1 MiB)  TX bytes:5731508707 (5.3 GiB)
eth0:0    Link encap:Ethernet  HWaddr 00:0C:29:1F:74:CF
          inet addr:172.16.22.100  Bcast:172.16.255.255  Mask:255.255.0.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
[root@jie2 ~]# ifconfig  eth0:0 down
[root@jie2 ~]# ifconfig | grep eth0
eth0      Link encap:Ethernet  HWaddr 00:0C:29:1F:74:CF
          inet addr:172.16.22.2  Bcast:172.16.255.255  Mask:255.255.0.0
          inet6 addr: fe80::20c:29ff:fe1f:74cf/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2165242 errors:0 dropped:0 overruns:0 frame:0
          TX packets:4110094 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:167917669 (160.1 MiB)  TX bytes:5731537035 (5.3 GiB)
[root@jie2 ~]# crm status
Last updated: Thu Aug  8 18:29:27 2013
Last change: Mon Sep  2 01:44:59 2013 via crm_attribute on jie3.com
Stack: classic openais (with plugin)
Current DC: jie3.com - partition with quorum
Version: 1.1.8-7.el6-394e906
2 Nodes configured, 2 expected votes
5 Resources configured.
Node jie3.com: standby
Online: [ jie2.com ]
 Master/Slave Set: ms_mysqldrbd [mysqldrbd]
     Masters: [ jie2.com ]
     Stopped: [ mysqldrbd:1 ]
 mystore    (ocf::heartbeat:Filesystem):    Started jie2.com
 myvip  (ocf::heartbeat:IPaddr):    Started jie2.com
 myserver   (lsb:mysqld):   Started jie2.com
Failed actions:
    myvip_monitor_20000 (node=jie2.com, call=47, rc=7, status=complete): not running
    myserver_monitor_20000 (node=jie3.com, call=209, rc=7, status=complete): not running
[root@jie2 ~]# ifconfig | grep eth0
eth0      Link encap:Ethernet  HWaddr 00:0C:29:1F:74:CF
          inet addr:172.16.22.2  Bcast:172.16.255.255  Mask:255.255.0.0
          inet6 addr: fe80::20c:29ff:fe1f:74cf/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:2165681 errors:0 dropped:0 overruns:0 frame:0
          TX packets:4110535 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:168015864 (160.2 MiB)  TX bytes:5731617112 (5.3 GiB)
eth0:1    Link encap:Ethernet  HWaddr 00:0C:29:1F:74:CF
          inet addr:172.16.22.100  Bcast:172.16.255.255  Mask:255.255.0.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
[root@jie2 ~]#


自此mysql的高可用已经完成。本博客没对crm命令参数进行解释