keepalived诞生的目的是为了给ipvs提供高可用性的.
keppalived 服务一般会启动两个进程,一个是vrrp服务和后端服务通信的,一个是checker服务,检测后端real Server健康状况.
邮件服务器:
rhel5:sendmail
rhel6:postfix
keepalived最新版本1.3.5,keepalived配置文件共三部分组成.
1
2
3
4
|
global_defs { #全局配置
notification_email { #收件人
main@example.com
} |
1
2
3
4
5
|
notification_email_from keepalived@admin #发件人
smtp_server 127.0.01 #发件服务器
smtp_connect_timeout #30超时时间
router_id nginx_slave #路由标识,自定义
} |
1
2
3
4
5
|
vrrp_script chk_port { #脚本检测名称chk_port
script "/etc/keepalived/keepalived.jk2.sh" #脚本的路径
interval 2 #每隔2秒检测一次
weight -2 #一旦失败,权重减2
} |
VRRP状态机,初始化(initialize)时,大家都是backup状态,通过选举产生master.收到startup且优先级是255时,直接定义为master,收到startup且优先级小雨255时,直接定义为backup.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
vrrp_instance VI_1 { #定义虚拟路由和虚拟ip的.VI_1为名称.
state MASTER interface eth0
virtual_router_id 51 #虚拟路由id,一般不大于255
priority 100 #初始优先级100,值越大优先级越高.
advert_int 1
authentication {
auth_type PASS #认证机制,明文认证
auth_pass 1111 #密码
}
virtual_ipaddress { #虚拟vip地址
192.168.30.129
}
track_script { #虚拟路由跟踪脚本.
chk_port
} } |
其他脚本定义使用
1
2
3
4
5
|
vrrp_script chk_file { script "[[-f /etc/keepalived/down]] && exit 1 || exit 0"
interval 1
weight -2
} |
实例:
系统Centos 6.5
2个node节点:
VIp:192.168.30.131
real server:192.168.30.129
real server:192.168.30.130
#两台real server 操作:
1
|
yum install nginx keepalived -y
|
#这两台real server 先配置好nginx,做static server.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
[root@haproxy keepalived] # cat /etc/nginx/conf.d/admin.conf
# # The default server # server { listen 80 default_server;
server_name _;
# Load configuration files for the default server block.
include /etc/nginx/default .d/*.conf;
location / {
root /data/www/ ;
index.htm index index.html index.php; }
} |
以示区别/data/www/index.html 亦两台real server 静态页面取ip地址最后1位.
#配置keepalived:
real server 192.168.30.129的keepalived配置文件:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
|
[root@web2 keepalived] # cat keepalived.conf
! Configuration File for keepalived
global_defs { notification_email {
215687833@qq.com #告警通知
}
notification_email_from keepalived@admin
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id nginx_slave #名字可以随便起,标示作用.
} vrrp_script chk_port { #检测脚本
script "/etc/keepalived/keepalived.jk2.sh"
interval 2 #每个2秒运行一次
weight -2 #失败,本机keepalived优先级减2
} vrrp_instance VI_1 { state BACKUP #初始化此节点为backup
interface eth0 #网卡eth0
virtual_router_id 51 #虚拟路由id
priority 100 #优先级,两台优先级可以是一样的,也可以一个高一个低.
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
192.168.30.131
}
track_script { chk_port } } |
real server 192.168.30.130的配置文件:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
|
[root@haproxy keepalived] # cat keepalived.conf
! Configuration File for keepalived
global_defs { notification_email {
215687833@qq.com }
notification_email_from keepalived@admin
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id nginx_master标示这是nginx master
} vrrp_script chk_port { script "/etc/keepalived/keepalived.jk2.sh"
interval 3
weight -2
} vrrp_instance VI_1 { state MASTER #初始化状态为master,两台real server都可以初始化为BACKUP状态,让它们之间自己选举.
interface eth0
virtual_router_id 51
priority 101 #优先级高于从节点
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
192.168.30.131
}
track_script { chk_port } } |
检测脚本路径和内容,赋予脚本可执行权限chmod a+x ...:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
|
[root@haproxy keepalived] # pwd
/etc/keepalived [root@haproxy keepalived] # cat keepalived.jk2.sh
#!/bin/bash ps -C nginx
if [[ $? - eq 0 ]]; then
exit 0
else /etc/init .d /nginx restart > /dev/null
sleep 3
ps -C nginx
if [[ $? - eq 0 ]]; then
exit 0
else
exit 1
fi
fi |
#此脚本主要判断本地nginx服务如果down 尝试启动1此,还是down就认为本节点下线,vip自动飘值bakcup节点.
#两台real server 启动keepalived服务:
1
|
# /etc/init.d/keepalived start |
查看keepalived 的log:
1
2
3
4
5
6
7
8
9
10
11
|
[root@haproxy conf.d] # tail -f /var/log/messages
Aug 4 14:56:09 haproxy Keepalived[51515]: Starting VRRP child process, pid=51518 Aug 4 14:56:09 haproxy Keepalived_vrrp[51518]: Netlink reflector reports IP 192.168.30.130 added Aug 4 14:56:09 haproxy Keepalived_healthcheckers[51517]: Netlink reflector reports IP 192.168.30.130 added Aug 4 14:56:09 haproxy Keepalived_healthcheckers[51517]: Netlink reflector reports IP fe80::20c:29ff:feca:1ae added Aug 4 14:56:09 haproxy Keepalived_healthcheckers[51517]: Registering Kernel netlink reflector Aug 4 14:56:09 haproxy Keepalived_healthcheckers[51517]: Registering Kernel netlink command channel
Aug 4 14:56:09 haproxy Keepalived_vrrp[51518]: Netlink reflector reports IP fe80::20c:29ff:feca:1ae added Aug 4 14:56:09 haproxy Keepalived_vrrp[51518]: Registering Kernel netlink reflector Aug 4 14:56:09 haproxy Keepalived_vrrp[51518]: Registering Kernel netlink command channel
Aug 4 14:56:09 haproxy Keepalived_vrrp[51518]: Registering gratuitous ARP shared channel |
#查看master的vip地址,ifconfig 看不到,用ip a或者:ip addr show
1
2
3
4
5
6
7
8
9
10
11
12
|
[root@haproxy conf.d] # ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link /loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1 /8 scope host lo
inet6 ::1 /128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000 link /ether 00:0c:29:ca:01:ae brd ff:ff:ff:ff:ff:ff
inet 192.168.30.130 /24 brd 192.168.30.255 scope global eth0
inet 192.168.30.131 /32 scope global eth0
inet6 fe80::20c:29ff:feca:1ae /64 scope link
valid_lft forever preferred_lft forever
|
#测试:打开浏览器访问http://192.168.30.131/ ,其中一台nginx 启动失败即可看到演示效果.