Nginx高可用-Keepalived
Keepalived:专门做高可用的软件,不只是做Nginx高可用
什么是高可用
一般是指2台机器启动着完全相同的业务系统,当有一台机器down机了,另外一台服务器就能快速的接管,对于访问的用户是无感知的
高可用通常使用什么软件
- keepalived
- heartbeat
- RoseHA
keepalived是如何实现高可用的?
keepalived软件是基于VRRP协议实现的,VRRP协议:虚拟路由冗余协议,主要用于解决单点故障问题
VRRP原理又是什么?
我们的VRRP其实是通过软件或者硬件的形式在Master和Backup外面增加一个虚拟的MAC地址(VMAC)与虚拟IP地址(VIP),那么在这种情况下,PC请求VIP的时候,无论是Master处理还是Backup处理,PC仅会在ARP缓存表中记录VMAC与VIP的信息。
高可用keepalived使用场景
通常业务系统需要保证7×24小时不DOWN机,比如公司内部的OA系统,每天公司人员都需要使用,则不允许Down机,作为业务系统来说随时都可用
高可用keepalived核心概
-
1、优先级
- 如何确定谁是主节点谁是备节点
-
2、抢占试、非抢占式
- 如果Master故障,Backup自动接管,那么Master回复后会夺权吗
-
3、脑裂
- 如果两台服务器都认为自己是Master会出现什么问题
部署keepalived高可用软件
环境准备
主机 | 角色 | 外网IP | 内网IP | 软件 |
---|---|---|---|---|
lb01 | 主节点(master) | 10.0.0.5 | 172.16.1.5 | nginx、keepalived |
lb02 | 备节点(backup) | 10.0.0.6 | 172.16.1.6 | nginx、keepalived |
VIP | 虚拟IP | 10.0.0.3 |
keeplived工作原理
1.哪些机器需要做高可用,就要在哪些机器上安装keepalived
2.keepalived的主节点会心跳检测(想要证明应用或者主机是否存活)
3.如果心跳检测失败,就杀掉自己(keepalived)
4.VIP到备节点
安装keepalived
# 1.安装keepalived软件
[root@lb01 ~]# yum install -y keepalived
[root@lb02 ~]# yum install -y keepalived
# 2.修改配置文件
## 修改master主节点配置文件
[root@lb01 ~]# vim /etc/keepalived/keepalived.conf
global_defs { #全局配置
router_id lb01 #标识身份->名称
}
vrrp_instance VI_1 {
state MASTER #标识角色状态
interface eth0 #网卡绑定接口
virtual_router_id 50 #虚拟路由id
priority 150 #优先级
advert_int 1 #监测间隔时间
authentication { #认证
auth_type PASS #认证方式
auth_pass 1111 #认证密码
}
virtual_ipaddress {
10.0.0.3 #虚拟的VIP地址
}
}
## 修改backup备节点配置文件
[root@lb02 ~]# vim /etc/keepalived/keepalived.conf
global_defs {
router_id lb02
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 50
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.3
}
}
# 3.启动keepalived服务并加入开机自启
[root@lb01 ~]# systemctl start keepalived.service
[root@lb01 ~]# systemctl enable keepalived.service
[root@lb02 ~]# systemctl start keepalived.service
[root@lb02 ~]# systemctl enable keepalived.service
# 4.查看服务是否开启
[root@lb01 ~]# ps -ef|grep [k]eepalive
root 7363 1 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
root 7364 7363 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
root 7365 7363 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
[root@lb02 ~]# ps -ef|grep [k]eepalive
root 7284 1 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
root 7285 7284 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
root 7286 7284 0 18:33 ? 00:00:00 /usr/sbin/keepalived -D
# 5.查看vip(虚拟IP)
[root@lb01 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:d5:75:e6 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.0.0.3/32 scope global eth0
valid_lft forever preferred_lft forever
# 6.检测vIP是否会漂移
[root@lb01 ~]# systemctl stop keepalived
[root@lb01 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:d5:75:e6 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
[root@lb02 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:a3:bc:3e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.6/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.0.0.3/32 scope global eth0
valid_lft forever preferred_lft forever
主节点和备节点的配置文件区别
Keepalived配置区别 | Master节点配置 | Backup节点配置 |
---|---|---|
route_id(唯一标识) | router_id lb01 | router_id lb02 |
state(角色状态) | state MASTER | state BACKUP |
priority(竞选优先级) | priority 150 | priority 100 |
抢占式配置文件
分为主节点和备节点,
主节点配置:state MASTER,
备节点配置:state BACKUP
## 修改master主节点配置文件
[root@lb01 ~]# vim /etc/keepalived/keepalived.conf
global_defs { #全局配置
router_id lb01 #标识身份->名称
}
vrrp_instance VI_1 {
state MASTER #标识角色状态
interface eth0 #网卡绑定接口
virtual_router_id 50 #虚拟路由id
priority 150 #优先级
advert_int 1 #监测间隔时间
authentication { #认证
auth_type PASS #认证方式
auth_pass 1111 #认证密码
}
virtual_ipaddress {
10.0.0.3 #虚拟的VIP地址
}
}
## 修改backup备节点配置文件
[root@lb02 ~]# vim /etc/keepalived/keepalived.conf
global_defs {
router_id lb02
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 50
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.3
}
}
非抢占是配置
1、两个节点的state都必须配置为BACKUP
2、两个节点都必须加上配置 nopreempt
3、其中一个节点的优先级必须要高于另外一个节点的优先级。 两台服务器都角色状态启用nopreempt后,必须修改角色状态统一为BACKUP,唯一的区分就是优先级。
[root@lb01 ~]# vim /etc/keepalived/keepalived.conf
global_defs {
router_id lb02
}
vrrp_instance VI_1 {
state BACKUP
nopreempt
interface eth0
virtual_router_id 50
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.3
}
}
[root@lb02 ~]# vim /etc/keepalived/keepalived.conf
global_defs {
router_id lb02
}
vrrp_instance VI_1 {
state BACKUP
nopreempt
interface eth0
virtual_router_id 50
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.3
}
}
keepalived结合nginx实现负载均衡的高可用
为什么域名解析到VIP就可以访问nginx?
Nginx默认监听在所有的IP地址上,VIP会飘到一台节点上,相当于那台nginx多了VIP这么一个网卡,所以可以访问到nginx所在机器
但是.....如果nginx宕机,会导致用户请求失败,但是keepalived没有挂掉不会进行切换,所以需要编写一个脚本检测Nginx的存活状态,如果不存活则kill掉keepalived
## 1.编写监听nginx的脚本
[root@lb01 ~]# vim check.sh
#!/bin/sh
nginx_count=$(ps -ef|grep [n]ginx|wc -l)
if [ $nginx_count -eq 0 ];then
systemctl stop keepalived
fi
#1.判断Nginx是否存活,如果不存活则尝试启动Nginx
#if [ $nginx_count -eq 0 ];then
# #systemctl start nginx
# #sleep 3
# #2.等待3秒后再次获取一次Nginx状态
# nginx_count=$(ps -ef|grep [n]ginx|wc -l)
# #3.再次进行判断, 如Nginx还不存活则停止Keepalived,让地址进行漂移,并退出脚本
# if [ $nginx_count -eq 0 ];then
# systemctl stop keepalived
# fi
#fi
## 2.修改主节点的配置文件
[root@lb01 ~]# vim /etc/keepalived/keepalived.conf
global_defs { #全局配置
router_id lb01 #标识身份->名称
}
## 注意:执行脚本的名字,最好不要带进程名
vrrp_script check_web {
# 脚本路径
script "/root/check.sh"
# 检测时间(每5秒执行一次检测脚本)
interval 5
}
vrrp_instance VI_1 {
state MASTER #标识角色状态
interface eth0 #网卡绑定接口
virtual_router_id 50 #虚拟路由id
priority 150 #优先级
advert_int 1 #监测间隔时间
authentication { #认证
auth_type PASS #认证方式
auth_pass 1111 #认证密码
}
track_script {
check_web
}
virtual_ipaddress {
10.0.0.3 #虚拟的VIP地址
}
}
## 3.重启keepalived
[root@lb01 ~]# systemctl restart keepalived
## 4.给脚本执行权限
[root@lb01 ~]# chmod +x /root/check.sh
## 5.检测负载均衡是否可以高可用
# 关闭nginx服务(检测vIP是否会漂移)
[root@lb01 ~]# systemctl stop nginx
[root@lb01 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:d5:75:e6 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
[root@lb02 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:a3:bc:3e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.6/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.0.0.3/32 scope global eth0
valid_lft forever preferred_lft forever
高可用keepalived故障脑裂
由于某些原因,导致两台keepalived高可用服务器在指定时间内,无法检测到对方的心跳,而此时的两台高可用服务器又都还活着。
脑裂故障原因
- 1、服务器网线松动等网络故障
- 2、服务器硬件故障发生损坏现象而崩溃
- 3、主备都开启firewalld防火墙
实现脑裂故障现象
# 将节点1和节点2的防火墙都打开,两台负载都有10.0.0.3IP,出现脑裂现象
[root@lb01 ~]# systemctl start firewalld
[root@lb01 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:d5:75:e6 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.0.0.3/32 scope global eth0
valid_lft forever preferred_lft forever
[root@lb02 ~]# systemctl start firewalld
[root@lb02 ~]# ip a
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
link/ether 00:0c:29:a3:bc:3e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.6/24 brd 10.0.0.255 scope global noprefixroute eth0
valid_lft forever preferred_lft forever
inet 10.0.0.3/32 scope global eth0
valid_lft forever preferred_lft forever
打开http://blog.dsr.com/,博客直接访问不了
解决脑裂故障方法
# 如果发生脑裂,则随机kill掉一台即可
# 在备节点上编写检测脚本, 测试如果能ping通主节点并且备节点还有VIP的话则认为产生了脑裂
[root@lb02 ~]# cat check_split_brain.sh
#!/bin/sh
vip=10.0.0.3
lb01_ip=10.0.0.5
while true;do
ping -c 2 $lb01_ip &>/dev/null
if [ $? -eq 0 -a `ip add|grep "$vip"|wc -l` -eq 1 ];then
echo "ha is split brain.warning."
else
echo "ha is ok"
fi
sleep 5
done