创建简单WEB高可用集群

1. 环境介绍

我这里弄了2个虚拟机，信息如下：

node1:192.168.168.201

node2:192.168.168.202

2.配置主机名

[root@node1 ~]# vim /etc/sysconfig/network           #编辑节点1主机名配置文件
#编辑
HOSTNAME=node1.linuxpanda.com
[root@node1 ~]# hostname node1.linuxpanda.com        #主机名立即生效

[root@node2 ~]# vim /etc/sysconfig/network          #编辑节点2主机名配置文件
#编辑
HOSTNAME=node2.linuxpanda.com
[root@node2 ~]# hostname node2.linuxpanda.com       #主机名立即生效

2.主机互信

具体可以参见我的上一篇博客：http://www.cnblogs.com/zhaojiedi1992/p/zhaojiedi_linux_023_sshgenkey.html

[root@node1 ~]# ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa         #生成rsa
[root@node1 ~]# ssh-copy-id -i  ~/.ssh/id_rsa.pub root@192.168.168.202  #复制公钥

[root@node2 ~]# ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa         #生成rsa
[root@node2 ~]# ssh-copy-id -i  ~/.ssh/id_rsa.pub root@192.168.168.201  #复制公钥

[root@node1 ~]# ssh 192.168.168.202 'ip addr show dev eth0 '    #测试node1 ssh node2

[root@node2 ~]# ssh 192.168.168.201 'ip addr show dev eth0 '    #测试node2 ssh node1

3.host文件配置

host配置文件配置主要用于名称解析， 我们不建议使用dns解析，否则dns无法解析就会导致集群服务的崩溃。

[root@node1 ~]# vim /etc/hosts
#添加如下2行
192.168.168.201 node1.linuxpanda.com node1
192.168.168.202 node2.linuxpanda.com node2
[root@node1 ~]# scp /etc/hosts root@node2:/etc/hosts      #host文件复制一份到node2中去
[root@node1 ~]# ping node2 -c 1 #node1 ping node2
[root@node2 ~]# ping node1 -c 1 #node2 ping node1

4.防火墙和selinux设置

[root@node1 ~]# service iptables stop                     #关闭node防火墙
iptables: Flushing firewall rules:                         [  OK  ]
iptables: Setting chains to policy ACCEPT: filter          [  OK  ]
iptables: Unloading modules:                               [  OK  ]
[root@node1 ~]# ssh node2 'service iptables stop'         #关闭node2防火墙
iptables: Flushing firewall rules: [  OK  ]
iptables: Setting chains to policy ACCEPT: filter [  OK  ]
iptables: Unloading modules: [  OK  ]

[root@node1 ~]# sed -i '/SELINUX/s/enforcing/disabled/' /etc/selinux/config  #修改node1selinux设置
[root@node1 ~]# setenforce 0                                                 #立马生效
[root@node1 ~]# ssh node2 "sed -i '/SELINUX/s/enforcing/disabled/'           #修改node1selinux设置
[root@node1 ~]# ssh node2 " setenforce 0'                                    #立马生效

5.时间同步

具体参见我的另一篇博客：http://www.cnblogs.com/zhaojiedi1992/p/zhaojiedi_linux_022_ntp.html

6.安装heartbeat

[root@node1 ~]#  yum  install heartbeat heartbeat-devel heartbeat-lib

[root@node2 ~]# ssh node2 'yum  install heartbeat heartbeat-devel heartbeat-lib'

7.编辑配置文件

[root@node1 ha.d]# cd /etc/ha.d/
[root@node1 ha.d]# cp /usr/share/doc/heartbeat-3.0./{authkeys,ha.cf,haresources} -p .
[root@node1 ha.d]# vim authkeys
#添加以下两行
auth
 md5 oracle

[root@node1 ha.d]# mv ha.cf ha.cf.bak
[root@node1 ha.d]# vim ha.cf
#添加如下行
logfile    /var/log/ha-log
keepalive 2
deadtime 30
warntime 10
initdead 120
bcast    eth0        # Linux
auto_failback on
node    node1.linuxpanda.com
node    node2.linuxpanda.com
ping 192.168.168.1

[root@node1 ha.d]# vim haresources
#添加如下行
node1.linuxpanda.com IPaddr::192.168.168.200//eth0 httpd

[root@node1 ha.d]# scp -p authkeys haresources ha.cf node2:/etc/ha.d/

8.安装httpd服务

[root@node1 ha.d]# yum install httpd
[root@node1 ha.d]# service httpd start
Starting httpd:                                            [  OK  ]
[root@node1 ha.d]# echo "node1" > /var/www/html/index.html
[root@node1 ha.d]# elinks node1 -source

[root@node2 ha.d]# yum install httpd
[root@node2 ha.d]# service httpd start
Starting httpd:                                            [  OK  ]
[root@node2 ha.d]# echo "node2" > /var/www/html/index.html
[root@node2 ha.d]# elinks node2 -source

[root@node1 ha.d]# for i in {..};do  ssh node$i 'service httpd stop';ssh node$i 'chkconfig httpd off' ; done;

9.启动heartbeat

[root@node1 ha.d]# for i in {..} ; do ssh node$i 'service heartbeat start' ; done;

10.测试

我们使用宿主机测试下：

[root@node1 ha.d]# ip addr show dev eth0                       #确认node1的ip信息，发现200ip资源在node1上分配的。
: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::c9:: brd ff:ff:ff:ff:ff:ff
    inet 192.168.168.201/ brd 192.168.168.255 scope global eth0
    inet 192.168.168.200/ brd 192.168.168.255 scope global secondary eth0
    inet6 fe80::20c:29ff:fec9:/ scope link
       valid_lft forever preferred_lft forever
[root@node2 ha.d]# ip addr show dev eth0                     #查看node2的ip信息
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:50:56:3f:42:13 brd ff:ff:ff:ff:ff:ff
    inet 192.168.168.202/24 brd 192.168.168.255 scope global eth0
    inet6 fe80::250:56ff:fe3f:4213/64 scope link 
       valid_lft forever preferred_lft forever

我们使用hb_standby脚本将node1转为standby，当然也是可以断点处理。

[root@node1 heartbeat]# /usr/share/heartbeat/hb_standby
Going standby [all].

查看日志信息：

[root@node1 heartbeat]# tail - /var/log/ha-log

May  :: node1.linuxpanda.com heartbeat: []: info: node1.linuxpanda.com wants to go standby [all]
May  :: node1.linuxpanda.com heartbeat: []: info: standby: node2.linuxpanda.com can take our all resources
May  :: node1.linuxpanda.com heartbeat: []: info: give up all HA resources (standby).
ResourceManager(default)[]:    //11_12:: info: Releasing resource group: node1.linuxpanda.com IPaddr::192.168.168.200//eth0 httpd
ResourceManager(default)[]:    //11_12:: info: Running /etc/init.d/httpd  stop
ResourceManager(default)[]:    //11_12:: info: Running /etc/ha.d/resource.d/IPaddr 192.168.168.200//eth0 stop
IPaddr(IPaddr_192.168.168.)[]:    //11_12:: INFO: IP status = ok, IP_CIP=
/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_192.168.168.200)[9196]:    2017/05/11_12:39:46 INFO:  Success
May  :: node1.linuxpanda.com heartbeat: []: info: all HA resource release completed (standby).
May  :: node1.linuxpanda.com heartbeat: []: info: Local standby process completed [all].
May  :: node1.linuxpanda.com heartbeat: []: WARN:  lost packet(s) for [node2.linuxpanda.com] [:]
May  :: node1.linuxpanda.com heartbeat: []: info: remote resource transition completed.
May  :: node1.linuxpanda.com heartbeat: []: info: No pkts missing from node2.linuxpanda.com!
May  :: node1.linuxpanda.com heartbeat: []: info: Other node completed standby takeover of all resources.

测试成功。