centos mariadb galera 多主同步 haprox keepalive 集群热备

环境说明：

131: haproxy keepalived

132: haproxy keepalived

134: mariadb galera

135: mariadb galera

1. mariadb galera 集群

查看是否安装

yum list installed mariadb*

卸载

yum remove mariadb*

安装

配置yum源

# MariaDB 10.1 CentOS repository list - created 2016-12-13 01:20 UTC
# http://downloads.mariadb.org/mariadb/repositories/
[mariadb]
name = MariaDB
baseurl = http://yum.mariadb.org/10.1/centos7-amd64
gpgkey=https://yum.mariadb.org/RPM-GPG-KEY-MariaDB
gpgcheck=1

安装mariadb

yum install mariadb-server mariadb-client

初始化数据库，在一台操作即可，其他会自动同步

#systemctl start mariadb

#mysql_secure_installation

设置mysql root密码，其他回答yes即可

配置galera 同步集群文件

关闭数据库，修改/etc/my.cnf.d/server.cnf

systemctl stop mariadb
vi /etc/my.cnf.d/server.cnf

[galera]
# Mandatory settings
wsrep_on=ON
wsrep_provider=/lib64/galera/libgalera_smm.so
wsrep_cluster_address="gcomm://192.168.65.128,192.168.65.129"
binlog_format=row
default_storage_engine=InnoDB
innodb_autoinc_lock_mode=2

以集群方式启动第一个mariadb

#galera_new_cluster

检查wsrep状态

mysql> MariaDB [(none)]> show status like "wsrep_%";

复制gelare 集群的配置文件到其他节点的server.cnf文件

启动mariadb

systemctl start mariadb

查看启动情况

systemctl status mariadb

2. 安装haproxy

解压安装

tar -xvzf haproxy-1.7.5.tar.gz

make  TARGET=linux3100 CPU=x86_64  PREFIX=/usr/local/haprpxy

make install PREFIX=/usr/local/haproxy

#数说明：
#TARGET=linux3100
#使用uname -r查看内核，如：2.6.18-371.el5，此时该参数就为linux26
#kernel 大于2.6.28的用：TARGET=linux2628
#CPU=x86_64   #使用uname -r查看系统信息，如x86_64 x86_64 x86_64 GNU/Linux，此时该参数就为x86_64
#PREFIX=/usr/local/haprpxy   #/usr/local/haprpxy为haprpxy安装路径

创建目录和文件

mkdir -p /usr/local/haproxy/conf #创建配置文件目录

mkdir -p /etc/haproxy #创建配置文件目录

touch /usr/local/haproxy/conf/haproxy.cfg #创建配置文件

ln -s /usr/local/haproxy/conf/haproxy.cfg /etc/haproxy/haproxy.cfg #添加配置文件软连接

cp -r /home/software/haproxy-1.7.5/examples/errorfiles /usr/local/haproxy/errorfiles #拷贝错误页面

ln -s /usr/local/haproxy/errorfiles /etc/haproxy/errorfiles #添加软连接

mkdir -p /usr/local/haproxy/log #创建日志文件目录

touch /usr/local/haproxy/log/haproxy.log #创建日志文件

ln -s /usr/local/haproxy/log/haproxy.log /var/log/haproxy.log #添加软连接

cp /home/software/haproxy-1.7.5/examples/haproxy.init /etc/rc.d/init.d/haproxy #拷贝开机启动文件

chmod +x /etc/rc.d/init.d/haproxy #添加脚本执行权限

chkconfig haproxy on #设置开机启动

ln -s /usr/local/haproxy/sbin/haproxy /usr/sbin #添加软连接

配置文件 haproxy.cfg

[root@yzb-centos72-2 conf]# vi haproxy.cfg

#全局配置
global
    #设置日志
    log 127.0.0.1 local1 info
    log 127.0.0.1 local2 warning
    chroot /usr/local/haproxy
    #用户与用户组
    user  root
    group root
    #守护进程启动
    daemon
    #最大连接数
    maxconn 4000

#默认配置
defaults
    log global
    mode http
#    option httplog
    option dontlognull
    timeout connect 5000
    timeout client 50000
    timeout server 50000

 listen admin-stats
        bind 0.0.0.0:1080 #监听端口  
        option httplog #采用http日志格式  
        stats refresh 30s #统计页面自动刷新时间  
        stats uri /stats #统计页面url  
        stats realm Haproxy Manager #统计页面密码框上提示文本  
        stats auth admin:admin #统计页面用户名和密码设置  
        #stats hide-version #隐藏统计页面上HAProxy的版本信息

listen mysql-cluster
    bind  *:5506
    mode    tcp
   # stats   enable
   # balance roundrobin
   # option  forwardfor
   # option  http-server-close
   # timeout http-keep-alive 500
    server  proxy1 172.20.4.135:3306 weight 5 check inter 2000
    server  proxy2 172.20.4.134:3306 weight 5 check inter 2000

service haproxy start #启动

service haproxy stop #关闭

service haproxy restart #重启

haproxy所在的IP为172.20.4.131

通过navicat连接

centos mariadb galera 多主同步 haprox keepalive 集群热备

连接后，我通过131、134、135任何连接操作表，数据均会保持同步。

我也尝试将135上的mariadb关掉后，通过131操作数据，均能正常，并且135上的mariadb启动后数据也会进行同步。

为了防止单点，同上在132上也配置了haproxy

将131上的haproxy拷贝的132

[root@yzb-centos72-2 software]# scp haproxy-1.7.5.tar.gz root@172.20.4.132:/home
root@172.20.4.132's password:
haproxy-1.7.5.tar.gz 100% 1703KB 1.7MB/s 00:00
[root@yzb-centos72-2 software]# ll

在132上解压，配置

tar -xvzf haproxy-1.7.5.tar.gz

mv haproxy-1.7.5 /usr/local

[root@yzb-centos72-3 local]# cd haproxy-1.7.5/

[root@yzb-centos72-3 haproxy-1.7.5]#make TARGET=linux3100 CPU=x86_64 PREFIX=/usr/local/haprpxy

[root@yzb-centos72-3 haproxy-1.7.5]#make install PREFIX=/usr/local/haproxy

[root@yzb-centos72-3 local]# cd haproxy

[root@yzb-centos72-3 haproxy]# mkdir conf

将131上的配置文件远程拷贝到132

[root@yzb-centos72-2 conf]#scp haproxy.cfg root@172.20.4.132:/usr/local/haproxy/conf
root@172.20.4.132's password:
haproxy.cfg 100% 1152 1.1KB/s 00:00
将132上的配置改为6606

listen mysql-cluster
    bind 0.0.0.0:6606
    mode    tcp
   # stats   enable
   # balance roundrobin
   # option forwardfor
   # option http-server-close
   # timeout http-keep-alive 500
    server proxy1 172.20.4.135:3306 weight 5 check inter 2000
    server proxy2 172.20.4.134:3306 weight 5 check inter 2000

校验配置文件

[root@yzb-centos72-3 sbin]# ./haproxy -f ../conf/haproxy.cfg -c
Configuration file is valid

启动 harpoxy

因为这台没有配置到系统服务，所以直接进入home启动文件

[root@yzb-centos72-3 sbin]# ./haproxy -f ../conf/haproxy.cfg
[root@yzb-centos72-3 sbin]# ps -ef | grep haproxy
root 12345 1 0 13:27 ? 00:00:00 ./haproxy -f ../conf/haproxy.cfg
root 12347 11905 0 13:27 pts/0 00:00:00 grep --color=auto haproxy

navicat客户端连接

centos mariadb galera 多主同步 haprox keepalive 集群热备

为haproxy配置日志输出

HAProxy 不会直接输出文件日志，需要借助 Linux 的 rsyslog 来让 HAProxy 输出日志

[root@yzb-centos72-3 ~]# vi /usr/local/haproxy/conf/haproxy.cfg

#全局配置
global
    #设置日志
    log 127.0.0.1 local1 info
    log 127.0.0.1 local2 warning
    chroot /usr/local/haproxy
    #用户与用户组
    user  root
    group root
    #守护进程启动
    daemon
    #最大连接数
    maxconn 4000

#默认配置
defaults
    log global
    mode http
#    option httplog
    option dontlognull
    timeout connect 5000
    timeout client 50000
    timeout server 50000

 listen admin-stats
        bind 0.0.0.0:1080 #监听端口  
        option httplog #采用http日志格式  
        stats refresh 30s #统计页面自动刷新时间  
        stats uri /stats #统计页面url  
        stats realm Haproxy Manager #统计页面密码框上提示文本  
        stats auth admin:admin #统计页面用户名和密码设置  
        #stats hide-version #隐藏统计页面上HAProxy的版本信息

listen mysql-cluster
    bind  *:6606
    mode    tcp
   # stats   enable
   # balance roundrobin
   # option  forwardfor
   # option  http-server-close
   # timeout http-keep-alive 500
    server  proxy1 172.20.4.135:3306 weight 5 check inter 2000
    server  proxy2 172.20.4.134:3306 weight 5 check inter 2000

意思是将 info 级（及以上）的日志推送到 rsyslog 的 local1 接口，将 warn 级（及以上）的日志推送到 rsyslog 的 local2 接口，并且所有 frontend 都默认使用 global 中的日志配置。

注意：info 级的日志会打印 HAProxy 处理的每一条请求，会占用很大的磁盘空间，在生产环境中，建议将日志级别调整为 notice。

为 rsyslog 添加 haproxy 日志的配置

[root@yzb-centos72-3 ~]# vi /etc/rsyslog.d/haproxy.conf
$ModLoad imudp
$UDPServerRun 514
$FileCreateMode 0644 #日志文件的权限
$FileOwner root #日志文件的owner
local1.* /var/log/haproxy.log #local1接口对应的日志输出文件
local2.* /var/log/haproxy_warn.log #local2接口对应的日志输出文件

修改 rsyslog 的启动参数

[root@yzb-centos72-3 ~]# vi /etc/sysconfig/rsyslog
# Options for rsyslogd
# Syslogd options are deprecated since rsyslog v3.
# If you want to use them, switch to compatibility mode 2 by "-c 2"
# See rsyslogd(8) for more details
SYSLOGD_OPTIONS="-c 2 -r -m 0"

重启 rsyslog
[root@yzb-centos72-3 ~]# service rsyslog restart

Redirecting to /bin/systemctl restart rsyslog.service

用 logrotate 进行日志切分

通过 rsyslog 输出的日志是不会切分的，所以需要通过 Linux 提供的 logrotate 来对日志文件进行切分。

使用 root 用户，创建 haproxy 日志切分配置文件：

[root@yzb-centos72-3 ~]# mkdir /root/logrotate

[root@yzb-centos72-3 ~]# vi /root/logrotate/haproxy

/var/log/haproxy.log /var/log/haproxy_warn.log {  #切分的两个文件名
    daily        #按天切分
    rotate 7
    create 0644 root root
    compress     #压缩旧日志
    delaycompress  #延迟一天压缩
    missingok    #忽略文件不存在的错误
    dateext      #旧日志加上日志后缀
    sharedscripts  #切分后的重启脚本只运行一次
    postrotate   #切分后运行脚本重载rsyslog，让rsyslog向新的日志文件中输出日志
    /bin/kill -HUP $(/bin/cat /var/run/syslogd.pid 2>/dev/null) &>/dev/null
    endscript
}

将 logrotate 配置在 crontab 中：

[root@yzb-centos72-3 logrotate]# vi /etc/crontab

SHELL=/bin/bash
PATH=/sbin:/bin:/usr/sbin:/usr/bin
MAILTO=root

# For details see man 4 crontabs

# Example of job definition:
# .---------------- minute (0 - 59)
# |  .------------- hour (0 - 23)
# |  |  .---------- day of month (1 - 31)
# |  |  |  .------- month (1 - 12) OR jan,feb,mar,apr ...
# |  |  |  |  .---- day of week (0 - 6) (Sunday=0 or 7) OR sun,mon,tue,wed,thu,fri,sat
# |  |  |  |  |
# *  *  *  *  * user-name  command to be executed
  0  0  *  *  * root       /usr/sbin/logrotate /root/logrotate/haproxy

重启haproxy后，就可以去查看日志了

[root@yzb-centos72-2 keepalived]# killall haproxy

[root@yzb-centos72-3 sbin]# cd /usr/local/haproxy/sbin/

[root@yzb-centos72-3 sbin]# ./haproxy -f ../conf/haproxy.cfg

[root@yzb-centos72-3 sbin]# tail -100f /var/log/haproxy.log

3. keepalived

下载解压

[root@yzb-centos72-2 software]# wget http://www.keepalived.org/software/keepalived-1.3.5.tar.gz

tar -xvzf keepalived-1.3.5.tar.gz

cd keepalived-.1.3.5

./configure --prefix=/usr/local/keepalived

若是报下面错误

configure: error:
!!! OpenSSL is not properly installed on your system. !!!
!!! Can not include OpenSSL headers files. !!!

需要安装openssl

# yum install openssl-devel

若是报下面警告

*** WARNING - this build will not support IPVS with IPv6. Please install libnl/libnl-3 dev libraries to support IPv6 with IPVS.

安装 libnl/libnl-3

yum install libnl*

[root@yzb-centos72-2 keepalived-1.3.5]# make && make install

使用 killall -0 检查 HAProxy 服务是否存在。如果没有 killall 命令，则需要安装 psmisc 包。

[root@yzb-centos72-3 keepalived]# yum install psmisc

注册为系统服务

# cp /usr/local/keepalived/sbin/keepalived /usr/sbin/
# cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
# touch /etc/init.d/keepalived
# chmod +x /etc/init.d/keepalived
# vi /etc/init.d/keepalived

/etc/init.d/keepalived 的文件内容：

#!/bin/sh
#
# Startup script for the Keepalived daemon
#
# processname: keepalived
# pidfile: /var/run/keepalived.pid
# config: /etc/keepalived/keepalived.conf
# chkconfig: - 21 79
# description: Start and stop Keepalived

# Source function library
. /etc/rc.d/init.d/functions

# Source configuration file (we set KEEPALIVED_OPTIONS there)
. /etc/sysconfig/keepalived

RETVAL=0

prog="keepalived"

start() {
    echo -n $"Starting $prog: "
    daemon keepalived ${KEEPALIVED_OPTIONS}
    RETVAL=$?
    echo
    [ $RETVAL -eq 0 ] && touch /var/lock/subsys/$prog
}

stop() {
    echo -n $"Stopping $prog: "
    killproc keepalived
    RETVAL=$?
    echo
    [ $RETVAL -eq 0 ] && rm -f /var/lock/subsys/$prog
}

reload() {
    echo -n $"Reloading $prog: "
    killproc keepalived -1
    RETVAL=$?
    echo
}

# See how we were called.
case "$1" in
    start)
        start
        ;;
    stop)
        stop
        ;;
    reload)
        reload
        ;;
    restart)
        stop
        start
        ;;
    condrestart)
        if [ -f /var/lock/subsys/$prog ]; then
            stop
            start
        fi
        ;;
    status)
        status keepalived
        RETVAL=$?
        ;;
    *)
        echo "Usage: $0 {start|stop|reload|restart|condrestart|status}"
        RETVAL=1
esac

exit $RETVAL

此处的/etc/init.d/keepalived就是原安装包中的文件，也可按如下方式操作

cp /home/software/keepalived-1.3.5/keepalived/etc/init.d/keepalived /etc/init.d/keepalived

chmod +x /etc/init.d/keepalived

因为我们使用非默认路径（/usr/local）安装keepalived, 故需要修改几处路径，以保证keepalived能正常启动, 需要修改的文件如下：

1.修改/etc/init.d/keepalived, 寻找大约15行左右的. /etc/sysconfig/keepalived, 修改为：. /usr/local/keepalived/etc/sysconfig/keepalived, 即指向正确的文件位置，同时在上述行下添加以下内容（将keepavlied主程序所在路径导入到环境变量PATH中）：

PATH="$PATH:/usr/local/keepalived/sbin"

export PATH

2. 修改/usr/local/keepalived/etc/sysconfig/keepalived文件，设置正确的服务启动参数
KEEPALIVED_OPTIONS="-D -f /usr/local/keepalived/etc/keepalived/keepalived.conf"

若按上面的这些设置就不用向/etc/sysconfig/ 拷贝对应的文件了。

编辑配置文件

[root@yzb-centos72-2 keepalived-1.3.5]# mkdir /etc/keepalived/
[root@yzb-centos72-2 keepalived-1.3.5]# cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/
[root@yzb-centos72-2 keepalived-1.3.5]# vi /etc/keepalived/keepalived.conf

global_defs {
    router_id LVS_DEVEL  #虚拟路由名称
}

#HAProxy健康检查配置
vrrp_script chk_haproxy {
    script "killall -0 haproxy"  #使用killall -0检查haproxy实例是否存在，性能高于ps命令
    interval 2   #脚本运行周期
    weight 2   #每次检查的加权权重值
}

#虚拟路由配置
vrrp_instance VI_1 {
    state MASTER           #本机实例状态，MASTER/BACKUP，备机配置文件中请写BACKUP
    interface eth0      #本机网卡名称，使用ifconfig命令查看
    virtual_router_id 51   #虚拟路由编号，主备机保持一致
    priority 100           #本机初始权重，备机请填写小于主机的值（例如100）
    advert_int 1           #争抢虚地址的周期，秒
    virtual_ipaddress {
        172.20.4.234      #虚地址IP，主备机保持一致
    }
    track_script {
        chk_haproxy        #对应的健康检查配置
    }
}

service keepalived start

service keepalived restart

service keepalived stop

也可用systemctl管理 keepalived服务

systemctl enable keepalived.service #设置开机自动启动

systemctl disable keepalived.service     #取消开机自动启动

systemctl start keepalived.service       #启动服务

systemctl restart keepalived.service     #重启服务

systemctl stop keepalived.service         #停止服务

systemctl status keepalived.service      #查看服务状态

设置防火墙打开

firewall-cmd --permanent–-add-service=keepalived

firewall-cmd --reload

再次强调：Keepalived的所有功能是配置keepalived.conf文件来实现的。

测试VIP飘逸

首先更改前面的一些配置

1. 将131、132上的haproxy 代理监听的端口5506、6606均改为6606.

2. 将131、132上的keepalived上的priority改为 99、100.

通过ifconfig    或 ip addr sh 可以看到绑定的虚拟IP

[root@yzb-centos72-3 sbin]# ip addr sh eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 06:69:a4:00:10:d2 brd ff:ff:ff:ff:ff:ff
    inet 172.20.4.132/24 brd 172.20.4.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet 172.20.4.234/32 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::469:a4ff:fe00:10d2/64 scope link
       valid_lft forever preferred_lft forever

查看132上的/var/log/message，看到keepalive master获取vip

May 10 15:00:55 localhost haproxy[10603]: Proxy mysql-cluster started.
May 10 15:00:56 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Script(chk_haproxy) succeeded
May 10 15:00:56 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Changing effective priority from 100 to 102
May 10 15:00:56 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) forcing a new MASTER election
May 10 15:00:57 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Transition to MASTER STATE
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Entering MASTER STATE
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) setting protocol VIPs.
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:01:01 yzb-centos72-3 systemd: Started Session 10408 of user root.
May 10 15:01:01 yzb-centos72-3 systemd: Starting Session 10408 of user root.

通过navicat 访问虚拟ip 172.20.4.234，连接haproxy的6606代理端口

centos mariadb galera 多主同步 haprox keepalive 集群热备

测试将master上haproxy kill掉

[root@yzb-centos72-3 sbin]# killall haproxy

查看132上的master上的message日志，看到监控的haproxy有问题，优先级从102降到了100，发现101高于当前的100，释放vip

ntos72-3 systemd: keepalived.service start operation timed out. Terminating.
May 10 14:58:48 yzb-centos72-3 systemd: Failed to start LVS and VRRP High Availability Monitor.
May 10 14:58:48 yzb-centos72-3 systemd: Unit keepalived.service entered failed state.
May 10 14:58:48 yzb-centos72-3 systemd: keepalived.service failed.
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Script(chk_haproxy) failed
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Changing effective priority from 102 to 100
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Received advert with higher priority101, ours 100
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Entering BACKUP STATE
May 10 14:59:48 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) removing protocol VIPs.
May 10 14:59:50 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 14:59:52 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 14:59:54 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 14:59:56 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 14:59:58 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1
May 10 15:00:00 yzb-centos72-3 Keepalived_vrrp[10355]: /usr/bin/killall -0 haproxy exited with status 1

这时查看131上的backup上的mesage，优先级99升到101，获取vip，变成master

May 10 14:58:59 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Script(chk_haproxy) succeeded
May 10 14:59:00 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Changing effective priority from 99 to

101
May 10 14:59:48 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) forcing a new MASTER election
May 10 14:59:49 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Transition to MASTER STATE
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Entering MASTER STATE
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) setting protocol VIPs.
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on eth0 for 172.20.4.234
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 14:59:50 yzb-centos72-2 Keepalived_vrrp[27012]: Sending gratuitous ARP on eth0 for 172.20.4.234

再次将原132上的master上的 haproxy启动，查看message日志，则从新获取vip

0355]: VRRP_Script(chk_haproxy) succeeded
May 10 15:00:56 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Changing effective priority from 100 to102
May 10 15:00:56 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) forcing a new MASTER election
May 10 15:00:57 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Transition to MASTER STATE
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Entering MASTER STATE
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) setting protocol VIPs.
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on eth0 for 172.20.4.234
May 10 15:00:58 yzb-centos72-3 Keepalived_vrrp[10355]: Sending gratuitous ARP on eth0 for 172.20.4.234

而刚才131的master有变成了backup

May 10 15:00:30 localhost haproxy[26346]: Connect from 10.6.226.106:60851 to 172.20.4.234:6606 (mysql-cluster/TCP)
May 10 15:00:56 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Received advert with higher priority102, ours 101
May 10 15:00:56 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) Entering BACKUP STATE
May 10 15:00:56 yzb-centos72-2 Keepalived_vrrp[27012]: VRRP_Instance(VI_1) removing protocol VIPs.
May 10 15:01:01 yzb-centos72-2 systemd: Started Session 10386 of user root.
May 10 15:01:01 yzb-centos72-2 systemd: Starting Session 10386 of user root.

再次尝试将132上的keepalived杀掉，

这131backup又变成了master，

则启动132上的keepalived后，131又变成了backup，132又变成了master。

也就是在131、132上的haproxy、keepalived都正常的情况下，132均会抢到vip，成为master，因为132的优先级102大于131上的101。

上面的测试过程中，navicat通过vip 234连接的数据不受影响。

至此，结束了^_^

参考：

http://zhangxugg-163-com.iteye.com/blog/1665419

http://www.thinksaas.cn/topics/0/778/778851.html

http://www.cnblogs.com/405845829qq/p/6773340.html

秒客网

centos mariadb galera 多主同步 haprox keepalive 集群热备

相关文章