大家知道keepalived会有四种状态的变化,每种状态变化时,都可以调用一个脚本
当进入Master状态时会呼叫notify_master
当进入Backup状态时会呼叫notify_backup
当发现异常情况时进入Fault状态呼叫notify_fault
当Keepalived程序终止时则呼叫notify_stop
进入Master和Backup这两种状态很容易理解了,就是分别变为主和从
进入Fault,简单的说就是keepalived发现自己有问题了,不能能再去参与Master竞选了,你得修好他才行。
进入这种状态一般是keepalived自身出问题了,或者keepalived检测的网卡出问题不通了,再或者就是我们自己写的检测业务的脚本返回错误
进入Stop 这个就容易理解了,执行service keepalived stop 或者 systemctl stop keepalived 就会进入这个状态。
设置selinux为宽松模式
# setenforce 0
# sed -i 's/^SELINUX=.*/SELINUX=permissive/g' /etc/selinux/config
CentOS防火墙默认是不允许keepalived使用 vrrp的组播。
如果不开启组播ip,keepalived双机不能实现热备的效果,只能实现负载的效果,即虚拟ip不能实现漂移 。
Check that the multicast IP and protocol for VRRP are allowed in the firewall on both servers.
For firewalld:
添加规则
# firewall-cmd --direct --permanent--add-rule ipv4 filterINPUT 0 --in-interface eth0 --destination 224.0.0.18 --protocol vrrp -j ACCEPT
# firewall-cmd --direct --permanent--add-ruleipv4 filterOUTPUT 0 --out-interface eth0 --destination 224.0.0.18 --protocol vrrp -j ACCEPT
# firewall-cmd --reload
删除规则
# firewall-cmd --direct --permanent--remove-ruleipv4 filterINPUT 0 --in-interface eth0 --destination 224.0.0.18 --protocol vrrp -j ACCEPT
# firewall-cmd --direct --permanent--remove-rule ipv4 filterOUTPUT 0 --out-interface eth0 --destination 224.0.0.18 --protocol vrrp -j ACCEPT
# firewall-cmd --reload
For iptables:
添加规则
# iptables -A INPUT -p vrrp -j ACCEPT
# iptables -A OUTPUT -p vrrp -j ACCEPT
# service iptables save
删除规则
# iptables -D INPUT -p vrrp -j ACCEPT
# iptables -D OUTPUT -p vrrp -j ACCEPT
# service iptables save
keepalived基本应用解析
http://blog.51cto.com/pangge/1301878
keepalived官方文档
http://www.keepalived.org/doc/introduction.html
使用keepalived实现redis的故障切换
http://peiqiang.net/2014/11/21/keepalived-and-redis.html
Kamailio High Availability Done Right with Keepalived
http://blog.unicsolution.com/2015/01/kamailio-high-availability-with.html
Keepalived Check and Notify Scripts