一、软件需求：
1.环境描述
操 作 系 统： CentOS 6.5 X64
MySQL数据库： MySQL 5.1.71.1
keepalived角色划分：
MASTER：10.8.1.11
BACKUP：10.8.1.12
VIP：10.8.1.13
2.系统包安装
1）gcc\keepalived相关
yum install -y pcre-devel openssl-devel popt-devel
2）MySQL相关
yum install -y mysql-server mysql
二、配置MySQL数据库Dual-Master
-
MySQL服务器配置
同一时刻只有一台keepalived角色MASTER STATE的MySQL数据库主机提供读写服务。
1）mdb01
[root@mdb01 ~]# more /etc/my.cnf
[mysqld]
server-id = 1
log-bin = mysql-bin
binlog-ignore-db = mysql,information_schema
auto-increment-increment = 2
auto-increment-offset = 1
slave-skip-errors = all
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
user=mysql
# Disabling symbolic-links is recommended to prevent assorted security risks
symbolic-links=0
[mysqld_safe]
log-error=/var/log/mysqld.log
pid-file=/var/run/mysqld/mysqld.pid
[root@mdb01 ~]#
2）mdb02
[root@mdb02 ~]# more /etc/my.cnf
[mysqld]
server-id = 2
log-bin = mysql-bin
binlog-ignore-db = mysql,information_schema
auto-increment-increment = 2
auto-increment-offset = 1
slave-skip-errors = all
datadir=/var/lib/mysql
socket=/var/lib/mysql/mysql.sock
user=mysql
# Disabling symbolic-links is recommended to prevent assorted security risks
symbolic-links=0
[mysqld_safe]
log-error=/var/log/mysqld.log
pid-file=/var/run/mysqld/mysqld.pid
[root@mdb02 ~]#
2.复制配置
1）复制用户设置
GRANT REPLICATION SLAVE ON *.* TO 'replication'@'10.8.1.%' IDENTIFIED BY 'replication';
flush privileges;
2）配置复制点
change master to
master_host='10.8.1.12',(Master IP)
master_user='replication',
master_password='replication',
master_log_file='mysql-bin.000001',(Master binlog)
master_log_pos=106; (Master binlog position)
3）启动复制进程
start slave;
4）查看复制状态
mysql> show slave status\G;
....
....
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
....
....
5)验证配置
在两个数据库端分别建立数据库、表，插入记录，如果在两边客户端均可见即成功。
三、安装配置keepalived
1.安装keepalived
[root@mdb01 keepalived-1.2.7]# tar zxvf keepalived-1.2.7.tar.gz
[root@mdb01 keepalived-1.2.7]# cd keepalived-1.2.7
[root@mdb01 keepalived-1.2.7]# ./configure --prefix=/usr/local/keepalived
....
....
Keepalived configuration
------------------------
Keepalived version : 1.2.7
Compiler : gcc
Compiler flags : -g -O2
Extra Lib : -lpopt -lssl -lcrypto
Use IPVS Framework : Yes
IPVS sync daemon support : Yes
IPVS use libnl : No
Use VRRP Framework : Yes
Use VRRP VMAC : Yes
SNMP support : No
Use Debug flags : No
[root@mdb01 keepalived-1.2.7]#
[root@mdb01 keepalived-1.2.7]# make
[root@mdb01 keepalived-1.2.7]# make install
2.复制相关文件
# cp /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d/
# cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
# mkdir /etc/keepalived/
# cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/
# cp /usr/local/keepalived/sbin/keepalived /usr/sbin/
3.设置启动项
启动优先级：启动MySQL之后启动keepalived
4.keepalived配置文件
1）mdb01
[root@mdb01 ~]# more /etc/keepalived/keepalived.conf
! Configuration File forkeepalived
global_defs {
notification_email {
test@sina.com
}
notification_email_from admin@test.com
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id MYSQL_HA
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 100
advert_int 1
nopreempt
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.8.1.13
}
}
virtual_server 10.8.1.13 3306 {
delay_loop 2
#lb_algo rr
#lb_kind DR
persistence_timeout 50
protocol TCP
real_server 10.8.1.11 3306 {
weight 3
notify_down /usr/local/keepalived/mysql.sh
TCP_CHECK {
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
[root@mdb01 ~]#
2）mdb02
[root@mdb02 ~]# more /etc/keepalived/keepalived.conf
! Configuration File forkeepalived
global_defs {
notification_email {
test@sina.com
}
notification_email_from admin@test.com
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id MYSQL_HA
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 90
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.8.1.13
}
}
virtual_server 10.8.1.13 3306 {
delay_loop 2
#lb_algo rr
#lb_kind DR
persistence_timeout 50
protocol TCP
real_server 10.8.1.12 3306 {
weight 3
notify_down /usr/local/keepalived/mysql.sh
TCP_CHECK {
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
[root@mdb02 ~]#
5.相关切换脚本
当检测到MySQL进程端口不存在时执行以下脚本mysql.sh，
停止本机keepalived进程及移除VIP，备机状态由Backup转换
Master State同时获取VIP。
[root@mdb01 ~]# more /usr/local/keepalived/mysql.sh
#!/bin/bash
#pkill keepalived（此行可能会出现VIP漂移过程中两台都有VIP的现象，杀进程不彻底）
/etc/init.d/keepalived stop
[root@mdb01 ~]#
6.启动keepalived
查看VIP绑定情况
四.切换测试
1.停止master state主机MySQL数据库
[root@mdb02 ~]# /etc/init.d/mysqld stop
Stopping mysqld: [ OK ]
2.mdb02移除VIP和停止keepalived进程
1）mdb02移除 protocol VIPs.
注意命令行输出中已无VIP显示
[root@mdb02 ~]# ip addr
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:ec:6e:29 brd ff:ff:ff:ff:ff:ff
inet 10.8.1.12/16 brd 10.8.255.255 scope global eth0
inet6 fe80::20c:29ff:feec:6e29/64 scope link
valid_lft forever preferred_lft forever
[root@mdb02 ~]#
Keepalived日志输出：
[root@mdb02 ~]# tail -f /var/log/messages
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: TCP connection to [10.8.1.12]:3306 failed !!!
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: Removing service [10.8.1.12]:3306 from VS [10.8.1.13]:3306
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: IPVS: Service not defined
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: Executing [/usr/local/keepalived/mysql.sh] for service [10.8.1.12]:3306 in VS [10.8.1.13]:3306
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: Lost quorum 1-0=1 > 0 for VS [10.8.1.13]:3306
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: Remote SMTP server [127.0.0.1]:25 connected.
Dec 1 22:01:05 mdb02 Keepalived[1118]: Stopping Keepalived v1.2.7 (12/01,2015)
Dec 1 22:01:05 mdb02 Keepalived_healthcheckers[1121]: IPVS: No such service
Dec 1 22:01:05 mdb02 Keepalived_vrrp[1122]: VRRP_Instance(VI_1) sending 0 priority
Dec 1 22:01:05 mdb02 Keepalived_vrrp[1122]: VRRP_Instance(VI_1) removing protocol VIPs.
2）keepalived进程消失
[root@mdb02 ~]# /etc/init.d/keepalived status
keepalived is stopped
3.mdb01 绑定VIP
[root@mdb01 ~]# ip addr
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:29:02:3a:68 brd ff:ff:ff:ff:ff:ff
inet 10.8.1.11/16 brd 10.8.255.255 scope global eth0
inet 10.8.1.13/32 scope global eth0
inet6 fe80::20c:29ff:fe02:3a68/64 scope link
valid_lft forever preferred_lft forever
4.mdb01转换为MASTER STATE
[root@mdb01 ~]# tail -f /var/log/messages
Dec 1 22:06:34 mdb01 Keepalived_vrrp[1137]: VRRP_Instance(VI_1) Transition to MASTER STATE
Dec 1 22:06:35 mdb01 Keepalived_vrrp[1137]: VRRP_Instance(VI_1) Entering MASTER STATE
Dec 1 22:06:35 mdb01 Keepalived_vrrp[1137]: VRRP_Instance(VI_1) setting protocol VIPs.
Dec 1 22:06:35 mdb01 Keepalived_vrrp[1137]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.8.1.13
Dec 1 22:06:35 mdb01 Keepalived_healthcheckers[1136]: Netlink reflector reports IP 10.8.1.13 added
Dec 1 22:06:40 mdb01 Keepalived_vrrp[1137]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 10.8.1.13
5.mdb02启动MySQL和Keepalived
MySQL和Keepalived服务均启动后，主机转换为BACKUP STATE
[root@mdb02 ~]# tail -f /var/log/messages
Dec 1 22:10:54 mdb02 Keepalived[2484]: Starting Keepalived v1.2.7 (12/01,2015)
Dec 1 22:10:54 mdb02 Keepalived[2485]: Starting Healthcheck child process, pid=2487
Dec 1 22:10:54 mdb02 Keepalived[2485]: Starting VRRP child process, pid=2488
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Interface queue is empty
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Netlink reflector reports IP 10.8.1.12 added
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Netlink reflector reports IP fe80::20c:29ff:feec:6e29 added
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Registering Kernel netlink reflector
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Registering Kernel netlink command channel
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Registering gratuitous ARP shared channel
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Opening file '/etc/keepalived/keepalived.conf'.
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Interface queue is empty
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Netlink reflector reports IP 10.8.1.12 added
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Netlink reflector reports IP fe80::20c:29ff:feec:6e29 added
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Registering Kernel netlink reflector
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Registering Kernel netlink command channel
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Opening file '/etc/keepalived/keepalived.conf'.
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Configuration is using : 62808 Bytes
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: Using LinkWatch kernel netlink reflector...
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Configuration is using : 11379 Bytes
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: VRRP_Instance(VI_1) Entering BACKUP STATE
Dec 1 22:10:54 mdb02 Keepalived_vrrp[2488]: VRRP sockpool: [ifindex(2), proto(112), fd(11,12)]
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: IPVS: Scheduler not found
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: IPVS: No such process
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Using LinkWatch kernel netlink reflector...
Dec 1 22:10:54 mdb02 Keepalived_healthcheckers[2487]: Activating healthchecker for service [10.8.1.12]:3306
Dec 1 22:10:54 mdb02 kernel: IPVS: Scheduler module ip_vs_ not found
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: TCP connection to [10.8.1.12]:3306 failed !!!
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: Removing service [10.8.1.12]:3306 from VS [10.8.1.13]:3306
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: IPVS: Service not defined
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: Executing [/usr/local/keepalived/mysql.sh] for service [10.8.1.12]:3306 in VS [10.8.1.13]:3306
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: Lost quorum 1-0=1 > 0 for VS [10.8.1.13]:3306
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: Remote SMTP server [127.0.0.1]:25 connected.
Dec 1 22:10:55 mdb02 Keepalived[2485]: Stopping Keepalived v1.2.7 (12/01,2015)
Dec 1 22:10:55 mdb02 Keepalived_healthcheckers[2487]: IPVS: No such service
VIP不做漂移
[root@mdb02 ~]# ip addr
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:ec:6e:29 brd ff:ff:ff:ff:ff:ff
inet 10.8.1.12/16 brd 10.8.255.255 scope global eth0
inet6 fe80::20c:29ff:feec:6e29/64 scope link
valid_lft forever preferred_lft forever
[root@mdb02 ~]#
五、总结：
正常情况下两台Master机器的MySQL及Keepalived均为启动状态：
1.当角色为MASTER STATE的MySQL数据库主机3306端口无法连接时即发生切换
1）停止keepalived进程。
2）移除VIP，
3）主机为故障状态，需要人工修复。
2.BACKUP STATE状态主机切换过程：
1）.转变角色为MASTER STATE。
向224.0.0.18发送VRRP心跳信息。
[root@mdb02 ~]# tcpdump vrrp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes
22:24:42.550593 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:43.563523 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:44.565224 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:45.567882 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:46.570211 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:47.572696 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:48.575324 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:49.578470 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:24:50.580935 IP mdb02 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
2）.绑定VIP，对外提供读写服务。
3.正常情况下，步骤1的故障主机修复后需要继续对外服务，启动需按照如下顺序：
1）启动MySQL数据库。
2）启动keepalived进程。
3）角色为BACKUP STATE。
4）接收vrrp心跳信息。
root@mdb01 ~]# tcpdump vrrp
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes
22:29:35.056846 IP 10.8.1.12 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:29:36.058852 IP 10.8.1.12 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
22:29:37.061641 IP 10.8.1.12 > 224.0.0.18: VRRPv2, Advertisement, vrid 51, prio 90, authtype simple, intvl 1s, length 20
5）不对外提供读写服务。
注意事项：
1.当两台服务器MySQL及Keepalived均为启动状态时，当发送故障切换，
VIP在故障机与备份机漂移需要适当时间间隔，一般10-20秒左右。
2.授权两台Mysql服务器允许root远程登录，用于在其他服务器登陆测试！
mysql> grant all on *.* to'root'@'10.8.1.%' identified by '123456';
mysql> flush privileges;
