Keepalived安装部署
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.
global_defs { #全局配置
notification_email { #收件人
main@example.com
}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
}
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.
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
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：
yum install nginx keepalived -y
#这两台real server 先配置好nginx，做static server.
# 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：
# 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：
# 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 ...：
# pwd
/etc/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：
# /etc/init.d/keepalived start
查看keepalived 的log：
# tail -f /var/log/messages
Aug4 14:56:09 haproxy Keepalived: Starting VRRP child process, pid=51518
Aug4 14:56:09 haproxy Keepalived_vrrp: Netlink reflector reports IP 192.168.30.130 added
Aug4 14:56:09 haproxy Keepalived_healthcheckers: Netlink reflector reports IP 192.168.30.130 added
Aug4 14:56:09 haproxy Keepalived_healthcheckers: Netlink reflector reports IP fe80::20c:29ff:feca:1ae added
Aug4 14:56:09 haproxy Keepalived_healthcheckers: Registering Kernel netlink reflector
Aug4 14:56:09 haproxy Keepalived_healthcheckers: Registering Kernel netlink command channel
Aug4 14:56:09 haproxy Keepalived_vrrp: Netlink reflector reports IP fe80::20c:29ff:feca:1ae added
Aug4 14:56:09 haproxy Keepalived_vrrp: Registering Kernel netlink reflector
Aug4 14:56:09 haproxy Keepalived_vrrp: Registering Kernel netlink command channel
Aug4 14:56:09 haproxy Keepalived_vrrp: Registering gratuitous ARP shared channel #查看master的vip地址,ifconfig 看不到，用ip a或者：ip addr show
# ip a
1: lo: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: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.
页:
[1]