简介
KeepAlived是网络协议VRRP（Virtual Route Redundancy Protocol 虚拟路由冗余协议）协议的实现。
KeepAlived是Linux集群管理中保证集群高可用的一个服务软件，其功能是用来防止单点故障。具体来说，其作用是检测服务器的状态，如果有一台web服务器宕机，或工作出现故障，KeepAlived将会检测到，并将有故障的服务器从系统中剔除，同时使用其他服务器代替该服务器的工作，当服务器工作正常后KeepAlived自动将服务器加入到服务器群中，这些工作全部自动完成，不需要人工干涉，需要人工做的只是修复故障的服务器。
KeepAlived的工作原理：
通过了解VRRP协议来了解KeepAlived的工作原理。
VRRP协议：Virtual Route Redundancy Protocol 虚拟路由冗余协议。是一种容错协议，保证当主机的下一跳路由出现故障时，由另一台备用路由器来代替出现故障的路由器进行工作，从而保持网络通信的连续性和可靠性。这些路由器组成了一个虚拟路由器，其中包含一个Master 路由器和多个 Backup 路由器。主机将虚拟路由器当作默认网关。一个虚拟路由器可以拥有一个或多个IP 地址。
实验实现基于KeepAlived双主模型的高可用LVS
需求分析
如下图所示
| FQDN | IP |
|---|---|
| images.king.com | VIP1 192.168.1.100 |
| app.king.com | VIP2 192.168.1.200 |
当客户端访问images.king.com的时候，是通过VIP1访问，最终后端提供服务的是WEBSRV1和WEBSRV2；当客户端访问app.king.com的时候，是通过VIP2访问，后端提供服务的是WEBSRV3和WEBSRV4。当用户访问其中一个站点的时候，单个LVS服务器可以提供负载均衡，将前端的请求调度到后端的两台WEBSRV上，但是一台LVS无法避免单点失败的问题，所以考虑用两台LVS服务器配合KeepAlived来实现高可用性。同时，企业提供两个站点的服务，如果每个站点都采用两台LVS来实现高可用性，那么需要四台LVS，成本比较高，而且正常情况下，有两台LVS服务器处于空闲中。可以考虑采用KeepAlived + LVS 双主模型的方式来实现这种需求。
双主模型。即针对VIP1，LVS1是MASTER，LVS2是BACKUP，此时，VIP1漂在LVS1上面，LVS1负责调度后端的WEBSRV1和WEBSRV2。针对VIP2，LVS2是MASTER，LVS1是BACKUP，此时，VIP2漂在LVS2上面，LVS2负责调度后端的WEBSRV3和WEBSRV4。当LVS1或者LVS2服务器出现故障的时候，那么此时出故障的LVS上面的VIP将漂移到另一台LVS上面，并且原来负责的两台后端WEBSRV将由正常的LVS接管。这就是大致的实现过程。
实验环境准备
确保所有虚拟机（centos7.3）上面的防火墙和selinux处于关闭状态。
各节点时间必须同步
找一台机器A与国内一台常用的NTP服务器同步时间，企业里面其他的机器与A机器同步时间。
各节点之间通过主机名互相通信，建议使用/etc/hosts文件实现
各节点之间的root用户可以基于密钥认证的ssh服务完成互相通信
实验过程
一、配置客户端CLIENT
vim /etc/hosts
# 添加下面两行
192.168.1.100 images.king.com
192.168.1.200 app.king.com
二、在WEBSRV1和WEBSRV2上面
WEBSRV2的配置和WEBSRV1的配置大致相同，下面以WEBSRV1为例介绍一下。
# 如果没有安装httpd就安装
yum install httpd
# 启动服务
systemctl start httpd
# 准备页面，如果是WEBSRV2，就把WEBSRV1 改为WEBSRV2
echo "WEBSRV1 images" > /var/www/html/index.html
# 准备RS脚本
vim lvs_dr_rs.sh
#!/bin/bash
vip=192.168.1.100
mask='255.255.255.255'
dev=lo:1
case $1 in
start)
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
ifconfig $dev $vip netmask $mask broadcast $vip up
route add -host $vip dev $dev
echo "The RS Server is Ready!"
;;
stop)
ifconfig $dev down
echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo "The RS Server is Canceled!"
;;
*)
echo "Usage: $(basename $0) start|stop"
exit 1
;;
esac
# 执行RS脚本
bash lvs_dr_rs.sh start
# 查看回环网卡上面是否绑定了vip 192.168.1.100
ip a
三、在WEBSRV3和WEBSRV4上面
WEBSRV4的配置和WEBSRV3的配置大致相同，下面以WEBSRV3为例介绍一下。
# 如果没有安装httpd就安装
yum install httpd
# 启动服务
systemctl start httpd
# 准备页面，如果是WEBSRV4，就把WEBSRV3 改为WEBSRV4
echo "WEBSRV3 app" > /var/www/html/index.html
# 准备RS脚本
vim lvs_dr_rs.sh
#!/bin/bash
vip=192.168.1.200
mask='255.255.255.255'
dev=lo:1
case $1 in
start)
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
ifconfig $dev $vip netmask $mask broadcast $vip up
route add -host $vip dev $dev
echo "The RS Server is Ready!"
;;
stop)
ifconfig $dev down
echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo "The RS Server is Canceled!"
;;
*)
echo "Usage: $(basename $0) start|stop"
exit 1
;;
esac
# 执行RS脚本
bash lvs_dr_rs.sh start
# 查看回环网卡上面是否绑定了vip 192.168.1.200
ip a
四、配置LVS1的KeepAlived
# 安装ipvsadm，用于查看ipvs规则
yum install ipvsadm
# 查看一下，目前来说是空的
ipvsadm -Ln
# 安装KeepAlived
yum install keepalived
# 配置keepalived
cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from node1@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id node1
vrrp_mcast_group4 224.100.100.100
}
vrrp_instance VI_1 {
state MASTER
interface ens33
virtual_router_id 88
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 411fa9f6
}
virtual_ipaddress {
192.168.1.100/24
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
virtual_server 192.168.1.100 80 {
delay_loop 3
lb_algo wrr
lb_kind DR
protocol TCP
sorry_server 127.0.0.1 80
real_server 192.168.1.14 80 {
weight 2
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 192.168.1.15 80 {
weight 1
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
vrrp_instance VI_2 {
state BACKUP
interface ens33
virtual_router_id 66
priority 80
advert_int 1
authentication {
auth_type PASS
auth_pass 123fa9f6
}
virtual_ipaddress {
192.168.1.200/24
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
virtual_server 192.168.1.200 80 {
delay_loop 3
lb_algo rr
lb_kind DR
protocol TCP
sorry_server 127.0.0.1 80
real_server 192.168.1.16 80 {
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 192.168.1.66 80 {
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
# MASTER和BACKUP切换的邮件通知脚本
cat /etc/keepalived/notify.sh
#!/bin/bash
contact='root@localhost'
notify() {
mailsubject="$(hostname) to be $1, vip floating"
mailbody="$(date +'%F %T'): vrrp transition, $(hostname) changed to be $1"
echo "$mailbody" | mail -s "$mailsubject" $contact
}
case $1 in
master)
notify master
;;
backup)
notify backup
;;
fault)
notify fault
;;
*)
echo "Usage: $(basename $0) {master|backup|fault}"
exit 1
;;
esac
# keepalived记录日志 ，修改/etc/sysconfig/keepalived对应的行
KEEPALIVED_OPTIONS="-D -S 2"
# 在/etc/rsyslog.conf添加一行
local2.* /var/log/keepalived.log
# 重启日志服务
systemctl restart rsyslog
# 准备sorry server
yum install httpd
systemctl start httpd
echo sorry,server > /var/www/html/index.html
# 启动keepalived服务
systemctl start keepalived
五、配置LVS2的KeepAlived
LVS2的配置大致和LVS1相同，主要是KeepAlived的配置不同，现将KeepAlived的配置说明如下。
cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from node2@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id node2
vrrp_mcast_group4 224.100.100.100
}
vrrp_instance VI_1 {
state BACKUP
interface ens33
virtual_router_id 88
priority 90
advert_int 1
authentication {
auth_type PASS
auth_pass 411fa9f6
}
virtual_ipaddress {
192.168.1.100/24
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
virtual_server 192.168.1.100 80 {
delay_loop 3
lb_algo wrr
lb_kind DR
protocol TCP
sorry_server 127.0.0.1 80
real_server 192.168.1.14 80 {
weight 2
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 192.168.1.15 80 {
weight 1
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
vrrp_instance VI_2 {
state MASTER
interface ens33
virtual_router_id 66
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 123fa9f6
}
virtual_ipaddress {
192.168.1.200/24
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
virtual_server 192.168.1.200 80 {
delay_loop 3
lb_algo rr
lb_kind DR
protocol TCP
sorry_server 127.0.0.1 80
real_server 192.168.1.16 80 {
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 192.168.1.66 80 {
HTTP_GET {
url {
path /index.html
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
六、在CLIENT上进行测试
测试一：
for i in {1..15};do curl images.king.com; done
for i in {1..15};do curl app.king.com; done
按照调度算法正常进行服务调度。
测试二：
for i in {1..50};do sleep 0.5; curl images.king.com;done
for i in {1..50};do sleep 0.5; curl app.king.com;done
断开LVS1或者LVS2的网络，两个站点依然正常访问。
此时，VIP1和VIP2集中在一台正常工作的LVS上面。
测试三：
for i in {1..50};do sleep 0.5; curl images.king.com;done
停掉WEBSRV1或者WEBSRV2，images.king.com站点依然可以正常访问。
for i in {1..50};do sleep 0.5; curl app.king.com;done
停掉WEBSRV3或者WEBSRV4，app.king.com站点依然可以正常访问。
测试四：
for i in {1..50};do sleep 0.5; curl images.king.com;done
把WEBSRV1和WEBSRV2的httpd服务都停掉，发现显示的是sorry,server，前端调度器充当了WEBSRV。
for i in {1..50};do sleep 0.5; curl app.king.com;done
把WEBSRV3和WEBSRV4的httpd服务都停掉，发现显示的是sorry,server，前端调度器充当了WEBSRV。