使用KeepAlived实现HAProxy高可用
在以前的文章中我们实现了使用HAProxy来转发读请求到后端多台mysql从库上，但是对于HAProxy的高可用(HA)需
求没有配置，本文在以前的基础再增加一个HAProxy的备份机器.这样当主服务器的HAProxy挂掉后，备份机器能够迅
速的进行自动接管.我们通过KeepAlived来实现这样的需求
主HAProxy: 192.168.10.192
备HAProxy: 192.168.10.85
VIP:192.168.10.158
整个项目的拓扑结构如下.
vip
|
--主haproxy-- --备haproxy--
| \
| \
--mysql-- --mysql--
--在主，备HAProxy上安装keepalived软件
[root@cscscslocalhost 1109]# tar -xvf keepalived-1.2.1.tar.gz
[root@cscscslocalhost keepalived-1.2.1]# ./configure --prefix=/usr/local/keepalived
checking for gcc... gcc
[root@cscscslocalhost keepalived-1.2.1]# make
[root@cscscslocalhost keepalived-1.2.1]# make install
--在主，备HAProxy上把KeepAlived的安装目录下的sbin目录加到.bash_profile文件中，然后source生效
PATH=$PATH:$HOME/bin:/usr/local/haproxy/sbin:/sbin:/bin:/usr/local/keepalived/sbin
--在主，备HAProxy上查看keepalived,确保成功安装.
[root@cscscslocalhost sbin]# keepalived -v
Keepalived v1.2.1 (11/12,)
--在主，备HAProxy上修改KeepAlived的配置文件
主HAProxy的配置
[root@cscscslocalhost keepalived]# vi /usr/local/keepalived/etc/keepalived/keepalived.conf
global_defs {
notification_email {
}
#notification_email_from
#smtp_server 192.168.200.1
#smtp_connect_timeout 30
router_id LVS_DEVEL
}
vrrp_script chk_haproxy
{
script "/usr/local/keepalived/chk_haproxy.sh"
interval 1
weight 2
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 51
priority 101
advert_int 1
mcast_src_ip 192.168.10.192
authentication {
auth_type PASS
auth_pass 111111
}
track_script {
chk_haproxy
}
virtual_ipaddress {
192.168.10.158
}
}
备用HAProxy服务器的配置
global_defs {
notification_email {
}
#notification_email_from
#smtp_server 192.168.200.1
#smtp_connect_timeout 30
router_id LVS_DEVEL
}
vrrp_script chk_haproxy
{
script "/usr/local/keepalived/chk_haproxy.sh"
interval 1
weight 2
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 100
advert_int 1
mcast_src_ip 192.168.10.85
authentication {
auth_type PASS
auth_pass 111111
}
track_script {
chk_haproxy
}
virtual_ipaddress {
192.168.10.158
}
}
--在主，备HAProxy上编写脚本文件，这个脚本的目的是检查HAProxy存活性，如果HAProxy
死掉了，那么KeepAlived也会被杀死，这样保证备HAProxy能确正常的接管
[root@cscscslocalhost ~]# vi /usr/local/keepalived/chk_haproxy.sh
#!/bin/bash
status=$(ps aux|grep haproxy | grep -v grep | grep -v bash | wc -l)
if [ ${status} = "0" ]; then
/bin/kill -9 `/sbin/pidof keepalived`
fi
--在主，备HAProxy上修改脚本执行权限
[root@cscscslocalhost ~]# chmod 755 /usr/local/keepalived/chk_haproxy.sh
--在主HAProxy上启动KeepAlived,通过/var/log/messages,可以看出已经功启动,并进入master状态
[root@cscscslocalhost keepalived]# keepalived -f /usr/local/keepalived/etc/keepalived/keepalived.conf
[root@cscscslocalhost ~]# tail -f /var/log/messages
Nov 12 14:28:15 cscscslocalhost Keepalived: Starting Keepalived v1.2.1 (11/12,)
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Registering Kernel netlink reflector
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Registering Kernel netlink command channel
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Registering gratutious ARP shared channel
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Opening file '/usr/local/keepalived/etc/keepalived/keepalived.conf'.
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Configuration is using : 64914 Bytes
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: Using LinkWatch kernel netlink reflector...
Nov 12 14:28:15 cscscslocalhost Keepalived: Starting VRRP child process, pid=17731
Nov 12 14:28:15 cscscslocalhost Keepalived_vrrp: VRRP_Script(chk_haproxy) succeeded
Nov 12 14:28:16 cscscslocalhost Keepalived_vrrp: VRRP_Instance(VI_1) Transition to MASTER STATE
Nov 12 14:28:17 cscscslocalhost Keepalived_vrrp: VRRP_Instance(VI_1) Entering MASTER STATE
[root@cscscslocalhost keepalived]# ip a
1: lo: mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
2: eth0: mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 1c:6f:65:d9:4a:6f brd ff:ff:ff:ff:ff:ff
inet 192.168.10.192/24 brd 192.168.10.255 scope global eth0
inet 192.168.10.158/32 scope global eth0
--测试通过VIP来连接mysql
[root@localhost ~]# mysql -u root -p123456 --host=192.168.10.158 --port=23306
Warning: Using a password on the command line interface can be insecure.
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 50484
Server version: 5.6.14-rel62.0 Percona Server with XtraDB (GPL), Release rel62.0, Revision 483
Copyright (c) - Percona LLC and/or its affiliates
Copyright (c) 2000, , Oracle and/or its affiliates. All rights reserved.
Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
mysql> show databases;
+--------------------+
| Database |
+--------------------+
| information_schema |
| db1 |
| mysql |
| performance_schema |
| login_sso |
| zabbix |
+--------------------+
6 rows in set (0.00 sec)
连接成功.
手动在主HAProxy上杀死haproxy,模拟haproxy故障,这样备份HAProxy会进行接管.
[root@cscscslocalhost keepalived]# ps -eaf | grep 4852
nobody 4852 1 0 15:13 ? 00:00:00 haproxy -f /etc/haproxy.cfg
root 5019 12171 0 15:13 pts/0 00:00:00 grep 4852
[root@cscscslocalhost keepalived]# kill -9 4852
--备份HAProxy的/var/log/messages日志,可以看出备份HAProxy已经接管了VIP
Nov 12 14:27:56 db135 Keepalived_vrrp: VRRP_Instance(VI_1) Received higher prio advert
Nov 12 14:27:56 db135 Keepalived_vrrp: VRRP_Instance(VI_1) Entering BACKUP STATE
Nov 12 14:28:41 db135 Keepalived_vrrp: VRRP_Instance(VI_1) Transition to MASTER STATE
Nov 12 14:28:42 db135 Keepalived_vrrp: VRRP_Instance(VI_1) Entering MASTER STATE
Nov 12 14:28:48 localhost.localdomain haproxy[12523]: Connect from 192.168.10.145:54535 to 192.168.10.158:23306 (mysql_proxy/TCP)
--重新show databases,发现已经重连了，整个过程在2-3秒的样子.
mysql> show databases;
No connection. Trying to reconnect...
Connection id: 50815
Current database: *** NONE ***
+--------------------+
| Database |
+--------------------+
| information_schema |
| db1 |
| mysql |
| performance_schema |
| login_sso |
| zabbix |
+--------------------+
6 rows in set (0.00 sec)
