heartbeat+ipvs+keepalived

1.高可用Heartbeat

Server1和server4安装资源包

   heartbeat-3.0.4-2.el6.x86_64.rpm

   heartbeat-devel-3.0.4-2.el6.x86_64.rpm

   heartbeat-libs-3.0.4-2.el6.x86_64.rpm

   ldirectord-3.9.5-3.1.x86_64.rpm

yum install -y heartbeat-*

[[email protected] ha.d]# cd /usr/share/doc/heartbeat-3.0.4/

[[email protected] heartbeat-3.0.4]# cp ha.cf authkeys haresources /etc/ha.d/

[[email protected] heartbeat-3.0.4]# cd /etc/ha.d

[[email protected] ha.d]# vim ha.cf

[[email protected] ha.d]# vim authkeys

[[email protected] ha.d]# chmod 600 authkeys 

[[email protected] ha.d]# vim haresources 

[[email protected] ha.d]# scp ha.cf authkeys haresources server4:/etc/ha.d

[[email protected] ha.d]# /etc/init.d/heartbeat start

Server4做相同的操作

[[email protected] ha.d]# /etc/init.d/heartbeat start

测试：

[[email protected] ha.d]# /etc/init.d/heartbeat stop

[[email protected] ha.d]# /etc/init.d/heartbeat start

1.ipvs负载均衡

首先停掉server1和server4上的heartbeat

Server1上的操作,并且server1做调度器

[[email protected] ~]# yum install -y ldirectord-3.9.5-3.1.x86_64.rpm

[[email protected] ~]# cd /usr/share/doc/ldirectord-3.9.5/

[[email protected] ldirectord-3.9.5]# cp ldirectord.cf /etc/ha.d

[[email protected] ~]# yum install -y ipvsadm

[[email protected] ~]# ipvsadm -L  

[[email protected] ~]# ipvsadm -C   ##清除规则

[[email protected] ~]# ipvsadm -A -t 172.25.6.100:80 -s rr    ##rr表示采用轮询调度算法，添加新的虚拟ip

[[email protected] ~]# ip addr add 172.25.6.100/24 dev eth0

[[email protected] ~]# ip addr

[[email protected] ~]# ipvsadm -a -t 172.25.6.100:80 -r 172.25.6.2:80 -g   ##添加新规则

[[email protected] ~]# ipvsadm -a -t 172.25.6.100:80 -r 172.25.6.3:80 -g

[[email protected] ~]# /etc/init.d/ipvsadm save   ##保存添加的规则

 [[email protected] ~]# ipvsadm -L      ##显示内核中虚拟服务器列表

[[email protected] ~]# /etc/init.d/httpd restart

Server2上的操作

[[email protected] ~]# ip addr add 172.25.6.100/32 dev eth0

[[email protected] ~]# ip addr

[[email protected] ~]# yum install -y arptables_jf

[[email protected] html]# arptables -A IN -d 172.25.6.100 -j DROP

[[email protected] html]# arptables -A OUT -s 172.25.6.200 -j mangle --mangle-ip-s 172.25.6.2

[[email protected] html]# /etc/init.d/arptables_jf  save

[[email protected] html]# /etc/init.d/httpd restart

Server3上的操作

[[email protected] ~]# /etc/init.d/httpd start

[[email protected] ~]# ip addr add 172.25.6.100/32 dev eth0   ##给内网网卡添加ip

[[email protected] ~]# ip addr

[[email protected] ~]# yum install -y arptables_jf

[[email protected] html]# arptables -A IN -d 172.25.6.100 -j DROP  ##给这个地址追加策略

[[email protected] html]# arptables -A OUT -s 172.25.6.100 -j  mangle --mangle-ip-s 172.25.6.3  ##追加策略，由100源地址跳到3上

[[email protected] ~]# /etc/init.d/arptables_jf save  ##保存添加的策略

[[email protected] ~]# /etc/init.d/arptables_jf start 

[[email protected] ~]# /etc/init.d/httpd restart

测试：

2.高可用与负载均衡

[[email protected] ~]# yum install -y ldirectord-3.9.5-3.1.x86_64.rpm

[[email protected] ~]# cd /etc/ha.d/

[[email protected] ha.d]# vim ldirectord.cf 

[[email protected] ha.d]# ipvsadm -C  ##清除策略

[[email protected] ha.d]# ipvsadm -L   ##查看策略

[[email protected] ha.d]# ip addr del 172.25.6.100/24 dev eth0    ##删除内网网卡上的ip

[[email protected] ha.d]# vim haresources 

[[email protected] ha.d]# scp haresources ldirectord.cf  172.25.6.4:/etc/ha.d/      ##server4上做与server1相同的操作

打开server4上的heartbeat替代server1的heartbeat

[[email protected] html]# /etc/init.d/httpd restart

测试：

如果server2和server3的httpd打开

如果server2和server3  httpd关闭，则访问heartbeat所在的主机

4.keepalived

关闭server1和server4的heartbeat，打开server2和server3的httpd

Server1上的操作

[[email protected] ~]# tar zxf keepalived-1.2.20.tar.gz

[[email protected] ~]#cd keepalived-1.2.20/

[[email protected] keepalived-1.2.20]# ./configure --prefix=/usr/local/keepalived/  ##解译到指定路径

[[email protected] keepalived-1.2.20]# make &&  make install

[[email protected] keepalived-1.2.20]# cd /usr/local/keepalived/

[[email protected] local]# scp -r keepalived/ 172.25.6.4:/usr/local/

[[email protected] sbin]#

 ln -s /usr/local/keepalived/sbin/keepalived /sbin/

[[email protected] etc]# ln -s /usr/local/keepalived/etc/keepalived/ /etc/

[[email protected] sysconfig]#

 ln -s /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/

[[email protected] sysconfig]#

 ln -s /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d/

[[email protected] init.d]# chmod +x keepalived

[[email protected] init.d]# cd /etc/keepalived/

[[email protected] keepalived]# ls

keepalived.conf  samples

[[email protected] keepalived]# vim keepalived.conf 

[[email protected] keepalived]# /etc/init.d/keepalived start

将修改的配置文件发送到server4同位置

Server4上的操作

[[email protected] keepalived]#

 ln -s /usr/local/keepalived/sbin/keepalived /sbin/

[[email protected] keepalived]#

 ln -s /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d

[[email protected] keepalived]#

 chmod +x /usr/local/keepalived/etc/rc.d/init.d/keepalived

[[email protected] keepalived]#

 ln -s /usr/local/keepalived/etc/sysconfig/keepalived /

etc/sysconfig/

[[email protected] keepalived]# /etc/init.d/keepalived start

[[email protected] keepalived]# mkdir /etc/keepalived/

将文件中的server1改为server4，state改为backup，

两者的virtual_router_id一定不同，server4的priority值小于server1的值

将keepalived进行重启

测试：