主机的ip地址分配

1）修改hostname

在node2和node3的/etc/hosts加入如下信息

192.168.1.120    node2.mictiger.com  node2
192.168.1.130   node3.mictiger.com  node3

2）建立master到node2，node3的节点互信

生成密钥对

[root@master ~]# ssh-keygen -t rsa -P ''    生成一对密钥
Generating public/private rsa key pair.
Enter file in which to save the key (/root/.ssh/id_rsa):
Your identification has been saved in /root/.ssh/id_rsa.
Your public key has been saved in /root/.ssh/id_rsa.pub.
The key fingerprint is:
dc:a5:7d:fc:ff:42:8c:8a:55:38:73:87:35:9a:f5:6a root@CentOS
The key's randomart p_w_picpath is:
+--[ RSA 2048]----+
|                 |
|              +  |
|           ..* o |
|       . .++*.. .|
|        S o=.+o. |
|          . ..E. |
|         o . o  .|
|        . .   . .|
|               .+|
+-----------------+

将生成的公钥传到目标主机

[root@master ~]# ssh-copy-id -i .ssh/id_rsa.pub root@192.168.1.120
[root@master ~]# ssh-copy-id -i .ssh/id_rsa.pub root@192.168.1.130
[root@master ~]# ssh-copy-id -i .ssh/id_rsa.pub root@192.168.1.150

在master上验证是否与node2，node3建立互信成功

[root@master ~]# ssh 192.168.1.130 'date';ssh 192.168.1.120 'date';date
Tue Nov 18 22:39:55 CST 2014
Tue Nov 18 22:40:03 CST 2014
Tue Nov 18 22:40:04 CST 2014

3）安装配置ansible

[root@master ~]# yum install ansible
[root@master ~]# vim /etc/ansible/hosts   只需要配置这文件配置要管理的主机
[HA]
192.168.1.150
192.168.1.120
192.168.1.130
[web]
192.168.1.120
192.168.1.130
[master]
192.168.1.150

验证ansible是否配置成功

[root@master ~]# ansible all -a 'date'
192.168.1.120 | success | rc=0 >>
Tue Nov 18 13:40:21 CST 2014192.168.1.150 | success | rc=0 >>
Tue Nov 18 13:40:21 CST 2014192.168.1.130 | success | rc=0 >>
Tue Nov 18 13:40:21 CST 2014

4）利用ansible更新时间，高可用群集必须配置一样的时间

[root@master ~]# ansible web -a 'ntpdate  202.118.1.81'
192.168.1.130 | success | rc=0 >>
18 Nov 22:50:54 ntpdate[21487]: step time server 202.118.1.81 offset 1.531700 sec192.168.1.120 | success | rc=0 >>
18 Nov 22:50:54 ntpdate[13616]: adjust time server 202.118.1.81 offset 0.313219 sec
将时间更新写入cron
[root@master ~]# ansible all -m cron -a 'name="sync time" minute="*/5" job="/usr/sbin/ntpdate 202.118.1.81 &> /dev/null"'
192.168.1.130 | success >> {"changed": true, "jobs": ["sync time"]
}192.168.1.120 | success >> {"changed": true, "jobs": ["sync time"]
}192.168.1.150 | success >> {"changed": true, "jobs": ["sync time"]
}

5) 安装并配置httpd程序

[root@master~]# ansible web -m yum -a 'name=httpd state=present'[root@master ~]# ansible web -a 'service httpd start'
192.168.1.130 | success | rc=0 >>
Starting httpd: [  OK  ]httpd: apr_sockaddr_info_get() failed for CentOS
httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName192.168.1.120 | success | rc=0 >>
Starting httpd: [  OK  ]httpd: Could not reliably determine the server's fully qualified domain name, using localhost.localdomain for ServerName
[root@master ~]# touch index.html
[root@master ~]# echo heartbeatV2 > index.html
[root@master ~]# scp index.html root@192.168.1.130:/var/www/html/
index.html                                                                                                                   100%   16     0.0KB/s   00:00
[root@master ~]# scp index.html root@192.168.1.120:/var/www/html/
index.html                                                                                                                   100%   16     0.0KB/s   00:00

验证httpd是否安装成功

[root@master ~]# ansible master -a 'curl  192.168.1.{120,130}'
192.168.1.150 | success | rc=0 >>
--_curl_--192.168.1.120
heartbeatV2
--_curl_--192.168.1.130
heartbeatV2
[1/2]: 192.168.1.120 --> <stdout>% Total    % Received % Xferd  Average Speed   Time    Time     Time  CurrentDload  Upload   Total   Spent    Left  Speed0    16    0    16    0     0   5072      0 --:--:-- --:--:-- --:--:--  8000[2/2]: 192.168.1.130 --> <stdout>0    16    0    16    0     0  10349      0 --:--:-- --:--:-- --:--:-- 10349

将httpd程序stop并不让它开机自动启动

[root@www ~]# ansible web -a 'service httpd stop'
192.168.1.120 | success | rc=0 >>
Stopping httpd: [  OK  ]192.168.1.130 | success | rc=0 >>
Stopping httpd: [  OK  ][root@www ~]# ansible web -a 'chkconfig httpd off'
192.168.1.120 | success | rc=0 >>192.168.1.130 | success | rc=0 >>

6）安装heatbeatV2需要处理依赖关系所以没有用ansible

[root@node2 ~]#  rpm -ivh heartbeat-2.1.4-12.el6.x86_64.rpm heartbeat-gui-2.1.4-12.el6.x86_64.rpm heartbeat-stonith-2.1.4-12.el6.x86_64.rpm heartbeat-pils-2.1.4-12.el6.x86_64.rpm
Preparing...                ########################################### [100%]1:heartbeat-pils         ########################################### [ 25%]2:heartbeat-stonith      ########################################### [ 50%]3:heartbeat              ########################################### [ 75%]4:heartbeat-gui          ########################################### [100%]

7）配置heartbeat

[root@node2 ~]# cp /usr/share/doc/heartbeat-2.1.4/{authkeys,ha.cf} /etc/ha.d/
配置authkeys
[root@node2 ha.d]# openssl rand -hex 16    生成16的随机数作为heartbeat传递心跳信息的验证
701782e0e6872f444edd0bbb726871fb
auth 2
#1 crc
#2 sha1 HI!
#3 md5 Hello!
2 sha1 701782e0e6872f444edd0bbb726871f
[root@CentOS ha.d]# chmod 600 authkeys    修改authkeys的权限不让其他人访问
配置heartbeat的主配置文件ha.cf大概配置如下信息（根据个人需求而定）
logfile /var/log/ha-log            日志文件位置
mcast eth0 225.10.10.10 694 1 0    使用组播地址通告和端口
auto_failback on                   自动添加down了又重新上线的节点到群集
node    node2.mictiger.com         配置节点信息
node    node3.mictiger.comping 192.168.1.1                   网关地址
配置haresources添加如下信息
node2.mictiger.com       192.168.1.250/24/eth0 httpd    主节点和VIP绑定的网卡，管理的资源
复制node2 的配置信息到node3
[root@localhost ~]# scp -r /etc/ha.d/* root@192.168.1.130:/etc/ha.d/

8）启动heartbeat

[root@master ~]# ansible web -a 'service heartbeat start'
192.168.1.130 | success | rc=0 >>
Starting High-Availability services:
Done.2014/11/18_23:36:19 INFO:  Resource is stopped192.168.1.120 | success | rc=0 >>
Starting High-Availability services:
Done.2014/11/18_23:36:19 INFO:  Resource is stopped

查看heartbeat运行在哪个节点上

[root@master ~]# ansible web -a 'ifconfig eth0:0'

192.168.1.130 | success | rc=0 >>

eth0:0    Link encap:Ethernet  HWaddr 00:0C:29:8F:A0:EF

UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

192.168.1.120 | success | rc=0 >>

eth0:0    Link encap:Ethernet  HWaddr 00:0C:29:57:5C:BB

inet addr:192.168.1.250  Bcast:192.168.1.255  Mask:255.255.255.0

UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

[root@master ~]# curl 192.168.1.250

heartbeatV2

关闭node2验证是否能够访问验证高可用性

[root@node2 ~]# service heartbeat stop

Stopping High-Availability services:

Done.