标签：script keepalived 双主
环境：
K1：192.168.11.26
K2：192.168.11.28
VIP1：192.168.11.17
VIP2：192.168.11.18
一、vrrp_script
1、K1配置
[root@K1 keepalived]# cat keepalived.conf #配置内容
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from xxx@163.com
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id LVS_DEVEL
# vrrp_mcast_group4 224.0.100.18
}
vrrp_script chk_down { #定义vrrp_script脚本扩展
script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" #判断文件是否存在，也可以定义脚本
interval 3 #检查周期
weight -10 #文件存在权重-10 根据上面script定义
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 6Nb6zjWB
}
virtual_ipaddress {
192.168.11.17
}
track_script { #跟踪script，如果不写，则不调用也就是不会执行我们是定义的vrrp_script定义的内容
chk_down
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}2、K2配置
[root@K2 keepalived]# cat keepalived.conf #配置内容
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from xxx@163.com
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id LVS_DEVEL
# vrrp_mcast_group4 224.0.100.18
}
vrrp_script chk_down { #定义vrrp_script脚本扩展
script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0" #判断文件是否存在，也可以定义脚本
interval 3 #检查周期
weight -10 #文件存在权重-10 根据上面script定义
}
}
vrrp_instance VI_1 {
state BAKCUP #填写BACKUP，因为K1是MASTER
interface eth0
virtual_router_id 51
priority 98 #优先级也需要设定比K1小
advert_int 1
authentication {
auth_type PASS
auth_pass 6Nb6zjWB
}
virtual_ipaddress {
192.168.11.17
}
track_script { #跟踪script，如果不写，则不调用也就是不会执行我们是定义的vrrp_script定义的内容
chk_down
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}3、测试
#启动keepalived服务，这里就不演示了 [root@K1 keepalived]# touch down #在K1MASTER上touche down文件 [root@K1 keepalived]# ls #查看是否存在，发现有down文件 down keepalived.conf keepalived.conf.bak keepalived.conf.real notify.sh [root@K1 keepalived]# tail -f /var/log/messages #在另外一个K1的session tail -f Jun 24 17:43:57 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) failed #发现我们定义的track_script（chk_down） failed ，自动权限-10 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Received higher prio advert #发现更高优先级的 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering BACKUP STATE #自己转为BACKUP STATE Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 17:43:58 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 [root@K1 keepalived]# rm -f down #删除down文件 You have new mail in /var/spool/mail/root #查看，down的确被删除 [root@K1 keepalived]# ls keepalived.conf keepalived.conf.bak keepalived.conf.real notify.sh [root@K1 keepalived]# tail -f /var/log/messages #切换回刚刚的session Jun 24 17:43:57 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) failed #发现我们定义的track_script（chk_down） failed ，自动权限-10，此时权限为90 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Received higher prio advert #发现更高优先级的 Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering BACKUP STATE #自己转为BACKUP STATE Jun 24 17:43:58 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议 Jun 24 17:43:58 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17 ========================================================割了，实际是连在一起的，只不过为了方便解释，割了============================================== Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Script(chk_down) succeeded #发现我们定义的tarck_script（chk_down） succeeded，此时权限为100，就是把之前-10的加回来了 Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER Jun 24 17:50:00 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MASTER Jun 24 17:50:01 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Transition to MASTER STATE #正在进入MASTER状态中 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Entering MASTER STATE #进入MASTER状态 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) setting protocol VIPs. #设定VIPs协议 Jun 24 17:50:02 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs Jun 24 17:50:02 web2 Keepalived_healthcheckers[9529]: Netlink reflector reports IP 192.168.11.17 added #添加192.168.11.17 Jun 24 17:50:07 web2 Keepalived_vrrp[9530]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs
4、通过vrrp_script简单检查服务并设定漂移
[root@K2 keepalived]# cat keepalived.conf #查看修改后的配置文件内容
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from 855@usfund.com.cn
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id LVS_DEVEL
# vrrp_mcast_group4 224.0.100.18
}
vrrp_script chk_down {
script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
interval 3
weight -10
}
vrrp_script chk_httpd { #增加一个新的vrrp_script，为chk_httpd
script "killall -0 httpd" #killall -0 信号，如果服务存在返回0
interval 3 #检查周期
weight -10 #httpd服务存在权重-10
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 6Nb6zjWB
}
virtual_ipaddress {
192.168.11.17
}
track_script {
chk_down
chk_httpd #增加了一个新的track_script chk_httpd
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
[root@K2 keepalived]# service keepalived stop #在K2停止keepalived
Stopping keepalived: [ OK ]
[root@K2 keepalived]# netstat -tnlp| grep httpd #查看httpd服务是否启动
tcp 0 0 :::80 :::* LISTEN 19187/httpd
[root@K2 keepalived]# service keepalived start #启动keepalived服务
Starting keepalived: [ OK ]
[root@K2 keepalived]# ip add sh #查看IP，发现：inet 192.168.11.17/32 scope global eth0
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:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff
inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.17/32 scope global eth0
inet6 fe80::20c:29ff:fe2d:bab0/64 scope link
valid_lft forever preferred_lft forever
[root@K1 keepalived]# service keepalived stop #在K1停止keepalived
正在启动 keepalived： [确定]
[root@K1 keepalived]# netstat -tunlp | grep httpd #查看httpd是否启动
tcp 0 0 :::80 :::* LISTEN 5457/httpd
[root@K1 keepalived]# service keepalived start #启动keepalived
[root@K1 keepalived]# ip add sh #查看IP，发现：192.168.11.17被抢占过来了
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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:ad:98:5c brd ff:ff:ff:ff:ff:ff
inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.17/32 scope global eth0
inet6 fe80::20c:29ff:fead:985c/64 scope link
valid_lft forever preferred_lft forever
[root@K1 keepalived]# service httpd stop #在K1停止httpd服务
停止 httpd： [确定]
You have new mail in /var/spool/mail/root
[root@K1 keepalived]# ip add sh #查看IP，发信192.168.11.17不见了
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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:ad:98:5c brd ff:ff:ff:ff:ff:ff
inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0
inet6 fe80::20c:29ff:fead:985c/64 scope link
valid_lft forever preferred_lft forever
[root@K1 keepalived]# tail -f /var/log/messages #在K1停止httpd服务器前，开启的一个新session
Jun 24 18:26:43 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) failed #定义的trackscript（chk_httpd）faileld
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Received higher prio advert #发现更高的优先级
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering BACKUP STATE #进入BACKUP状态
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议
Jun 24 18:26:44 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17
[root@K1 keepalived]# service httpd start #启动httpd服务
正在启动 httpd：httpd: apr_sockaddr_info_get() failed for K1
httpd: Could not reliably determine the server‘s fully qualified domain name, using 127.0.0.1 for ServerName
[确定]
You have new mail in /var/spool/mail/root
[root@K1 keepalived]# ip add sh #查看IP，发现：192.168.11.17被抢占过来了
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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:ad:98:5c brd ff:ff:ff:ff:ff:ff
inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.17/32 scope global eth0
inet6 fe80::20c:29ff:fead:985c/64 scope link
valid_lft forever preferred_lft forever
[root@K1 keepalived]#tail -f /var/log/messages #在K1停止httpd服务器前，开启的一个新session
Jun 24 18:26:43 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) failed #定义的trackscript（chk_httpd）faileld
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Received higher prio advert #发现更高的优先级
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering BACKUP STATE #进入BACKUP状态
Jun 24 18:26:44 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) removing protocol VIPs. #迁移VIPs协议
Jun 24 18:26:44 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 removed #宣布删除IP 192.168.11.17
========================================================割了，实际是连在一起的，只不过为了方便解释，割了==============================================
Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Script(chk_httpd) succeeded #发现我们定义的tarck_script（chk_httpd） succeeded，此时权限为100，就是把之前-10的加回来了
Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER
Jun 24 18:31:10 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) forcing a new MASTER election #推选了一个新的MATER
Jun 24 18:31:11 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Transition to MASTER STATE #正在进入MASTER状态中
Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Entering MASTER STATE #进入MASTER状态
Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) setting protocol VIPs. #设定VIPs协议
Jun 24 18:31:12 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs
Jun 24 18:31:12 web2 Keepalived_healthcheckers[11700]: Netlink reflector reports IP 192.168.11.17 added #添加192.168.11.17
Jun 24 18:31:17 web2 Keepalived_vrrp[11701]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 #从eth0发送192.168.11.17免费的ARPs二、双主模型
[root@K2 keepalived]# cat keepalived.conf #查案配置内容，K1的就不贴上来了，新增段配置改为相反的
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from xxx@163.com
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id LVS_DEVEL
# vrrp_mcast_group4 224.0.100.18
}
vrrp_script chk_down {
script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
interval 3
weight -10
}
vrrp_script chk_httpd {
script "killall -0 httpd"
interval 3
weight -10
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 6Nb6zjWB
}
virtual_ipaddress {
192.168.11.17
}
track_script {
chk_down
chk_httpd
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
vrrp_instance VI_2 { #相对上面的复制了VI_1内容并改为VI_2,配置也做了相应的修改
state MASTER #由复制BACKUP改为MASTER
interface eth0
virtual_router_id 52 #由51改为52
priority 100 #由98改为100
advert_int 1
authentication {
auth_type PASS
auth_pass 6Nb6zjWB
}
virtual_ipaddress {
192.168.11.18 #地址192.168.11.17改为192.168.11.18
}
track_script {
chk_down
chk_httpd
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
[root@K2 keepalived]# service keepalived stop #停止keepalived服务
Stopping keepalived: [ OK ]
[root@K2 keepalived]# service keepalived start #启动keepalived服务【需要等待两边都stop keepalived 服务，不然会出现一些小问题】
[root@K2 keepalived]# ip add sh #查看VIP地址都在这台，因为另外一台没有开启
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:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff
inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.18/32 scope global eth0
inet 192.168.11.17/32 scope global eth0
inet6 fe80::20c:29ff:fe2d:bab0/64 scope link
valid_lft forever preferred_lft forever
[root@K1 keepalived]# service keepalived stop #停止keepalived服务
停止 keepalived： [确定]
[root@K1 keepalived]# service keepalived start #启动keepalived服务
正在启动 keepalived： [确定]
[root@K1 keepalived]# ip add sh #哈哈，我胡三又回来了（192.168.11.17）
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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:ad:98:5c brd ff:ff:ff:ff:ff:ff
inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.17/32 scope global eth0
inet6 fe80::20c:29ff:fead:985c/64 scope link
valid_lft forever preferred_lft forever
[root@K2 keepalived]# ip add sh #在K2查看，192.168.11.17 没了
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:29:2d:ba:b0 brd ff:ff:ff:ff:ff:ff
inet 192.168.11.28/24 brd 192.168.11.255 scope global eth0
inet 192.168.11.18/32 scope global eth0
inet6 fe80::20c:29ff:fe2d:bab0/64 scope link
valid_lft forever preferred_lft forever
You have new mail in /var/spool/mail/root
#测试就不测试了 ^-^ 不管是httpd访问停止还是在/etc/keepalived/下touch down 服务都会跑到另外一边去本文出自 “SunshineBoySZF” 博客，请务必保留此出处http://sunshineboyszf.blog.51cto.com/12087328/1864488
标签：script keepalived 双主
原文地址：http://sunshineboyszf.blog.51cto.com/12087328/1864488
