lvs-nat 模式-piranha

系统: redhat 6.5 mini

1、  在各个机器上配置ip地址，机器名，关闭selinux,同步时间

[[email protected] ~]# hostname

lvs01

[[email protected] ~]# getenforce

Disabled

[[email protected] ~]# hostname

lvs02

[[email protected] ~]# getenforce

Disabled

[[email protected] ~]# hostname

rs01

[[email protected] ~]# getenforce

Disabled

[[email protected] ~]# hostname

rs02

[[email protected] ~]# getenforce

Disabled

[[email protected] yum.repos.d]# cat /etc/hosts

192.168.20.10 lvs01

192.168.20.20 lvs02

192.168.20.100 rs01

192.168.20.200 rs02

[[email protected] yum.repos.d]# ntpdate 0.rhel.pool.ntp.org

[[email protected] yum.repos.d]# ntpdate 0.rhel.pool.ntp.org

[[email protected] yum.repos.d]# ntpdate 0.rhel.pool.ntp.org

[[email protected] yum.repos.d]# ntpdate 0.rhel.pool.ntp.org

2、  在各个机器关闭防火墙，配置yum源

[[email protected] yum.repos.d]# service iptables stop

[[email protected] yum.repos.d]# chkconfig iptables off

[[email protected] ~]# service iptables stop

[[email protected] ~]# chkconfig iptables off

[[email protected] ~]# service iptables stop

[[email protected] ~]# chkconfig iptables off

[[email protected] ~]# service iptables stop

[[email protected] ~]# chkconfig iptables off

[[email protected] yum.repos.d]# service iptables stop

[[email protected] yum.repos.d]# chkconfig iptables off

[[email protected] yum.repos.d]# cat rhel-source.repo

[base]

name=base

baseurl=file:///mnt

enabled=1

gpgcheck=0

[LB]

name=LB

baseurl=file:///mnt/LoadBalancer

enabled=1

gpgcheck=0

[HA]

name=HA

baseurl=file:///mnt/HighAvailability

enabled=1

gpgcheck=0

[RS]

name=RS

baseurl=file:///mnt/ResilientStorage

enabled=1

gpgcheck=0

[[email protected] yum.repos.d]# scp rhel-source.repo lvs02:/etc/yum.repos.d/rhel-source.repo

3、  在rs01.rs02上部署web服务

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

[[email protected] ~]# echo "rs01">/var/www/html/index.html

[[email protected] ~]# service httpd restart

[[email protected] ~]# chkconfig httpd on

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

[[email protected] ~]# echo "rs02">/var/www/html/index.html

[[email protected] ~]# service httpd restart

[[email protected] ~]# chkconfig httpd on

4、  在lvs01,lvs02安装LB，开启路由转发功能，在lvs01上配置lvs集群登录用户和密码

[[email protected] yum.repos.d]# yum groupinstall "Load Balancer"

[[email protected] yum.repos.d]# yum groupinstall "Load Balancer"

[[email protected] yum.repos.d]# vi /etc/sysctl.conf

net.ipv4.ip_forward = 1

[[email protected] yum.repos.d]# sysctl –p

[[email protected] yum.repos.d]# vi /etc/sysctl.conf

net.ipv4.ip_forward = 1

[[email protected] yum.repos.d]# sysctl –p

[[email protected] yum.repos.d]# piranha-passwd

New Password:

Verify:

Adding password for user piranha

[[email protected] yum.repos.d]# /etc/init.d/piranha-gui start

[[email protected] yum.repos.d]# chkconfig piranha-gui on

5、  通过浏览器来配置lvs集群

5.1 在浏览器输入http://192.168.10.10:3636,点击login,输入用户名和密码

5.2 配置lvs01，选择nat 模式，最后点击accept

5.3 配置lvs02的ip，先点击ENABLE,配置ip 后，最后点击accept确定

5.4 配置对外的服务，点击add->edit->accept

5.5 配置真实服务器（rs01和rs02）步骤add->edit->accept->active

5.6  up对外服务的vip

5.7 将在lvs01 上的配置文件复制到lvs02 上，

[[email protected] ~]# scp /etc/sysconfig/ha/lvs.cf lvs02:/etc/sysconfig/ha/lvs.cf

[email protected]‘s password:

lvs.cf                              100%  865     0.8KB/s   00:00

[[email protected] ~]# cat /etc/sysconfig/ha/lvs.cf

serial_no = 17

primary = 192.168.10.10

primary_private = 192.168.20.10

service = lvs

backup_active = 1

backup = 192.168.10.20

backup_private = 192.168.20.20

heartbeat = 1

heartbeat_port = 539

keepalive = 6

deadtime = 18

network = nat

nat_router = 192.168.20.254 eth1:0

nat_nmask = 255.255.255.0

debug_level = NONE

monitor_links = 0

syncdaemon = 0

virtual http {

active = 1

address = 192.168.10.254 eth0:1

vip_nmask = 255.255.255.0

port = 80

send = "GET / HTTP/1.0\r\n\r\n"

expect = "HTTP"

use_regex = 0

load_monitor = none

scheduler = rr

protocol = tcp

timeout = 6

reentry = 15

quiesce_server = 0

server rs01 {

address = 192.168.20.100

active = 1

weight = 1

}

server rs02 {

address = 192.168.20.200

active = 1

weight = 2

}

}

5.8 在lvs01,lvs02 上启动负载均衡的监控进程

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

Shutting down pulse:                                       [失败]

Starting pulse:                                            [确定]

[[email protected] ~]# chkconfig pulse on

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

Shutting down pulse:                                       [失败]

Starting pulse:                                            [确定]

[[email protected] ~]# chkconfig pulse on

5.9 查看lvs的正常运行状态

6、  测试

6.1 访问vip，web的内容是后台两台真实服务器网页内容，轮替提供服务

６.2　lvs01失效的情况下服务依然可用（我直接把lvs01关机了）

[[email protected] ~]# elinks -dump http://192.168.10.254

rs01

原文地址：https://www.cnblogs.com/derrickrose/p/8542364.html