《童虎学习笔记》5分钟用Keepalived搭建MySQL高可用实现虚ip自动漂移

本文章配套视频	https://www.ixigua.com/7079764967548781086?id=7082376934071468580
本专栏全部文章	https://blog.csdn.net/tonghu_note/category_11716546.html
总目录	https://blog.csdn.net/tonghu_note/article/details/124333034

来我的dou音 aa10246666， 看配套视频

---

一、实战环境

主节点 node1 mysql 8.0.28	10.211.55.9
从节点 node2 mysql 8.0.28	10.211.55.4
Keepalived 2.1.5	10.211.55.100（VIP）

本节课程前提条件是node1(master)和node2(slave)的主从架构已经搭建完毕，如果你不知道如何搭建主从，那么可以参考这篇文章 《童虎学习笔记》5分钟搭建MySQL主从复制_童虎学习笔记的博客-CSDN博客童虎学习笔记的博客_CSDN博客-领域博主https://blog.csdn.net/tonghu_note来我的西瓜视频，看免费配套视频https://www.ixigua.com/home/2058760810138187来我的dou音 aa10246666， 看配套视频一、实战环境主节点 node1 mysql 810.211.55.9从节点 node2 mysql 810.211.55.4二、搭建主从1、主库新建主从同步帐号repl...https://blog.csdn.net/tonghu_note/article/details/123808841

---

二、编写mysql存活检查脚本

主从上都要有这个脚本，用于判断mysql进程是否存在，脚本内容如下

root@node1:/usr/local/scripts# cat chk_mysql_alived.sh 

#!bin/bash

ss -tnl|grep 3306 >/dev/null 2>&1

if [ $? -eq 0 ]
then
    echo " mysql is alived "  
    exit 0
else
    echo " mysql is dead "
    systemctl stop keepalived
    exit 2
fi

授权执行权限

chmod u+x /usr/local/scripts/chk_mysql_alived.sh

执行一下脚本试试效果

root@node1:/usr/local/scripts# sh chk_mysql_alived.sh 
 mysql is alived 
root@node1:/usr/local/scripts# echo $?
0

---

 三、安装及配置Keepalived

1、主从服务器上都安装Keepalived

apt-get install -y ipset
apt-get install -y keepalived

如果你用的是centos，可以用yum安装

2、主库服务器上部署Keepalived

root@node1:~# cat /etc/keepalived/keepalived.conf

vrrp_script chk_mysql_alived {
  script "/usr/local/scripts/chk_mysql_alived.sh"
  interval 2 
  weight 2
}

global_defs {
  router_id LVS_DEVEL_1
}

vrrp_instance VI_1 {
  state BACKUP 
  interface enp0s5
  track_interface {
    enp0s5
  }
  virtual_router_id 58
  priority 200
  advert_int 1
  nopreempt
  authentication {
    auth_type PASS
    auth_pass 1111
  }
  track_script {
    chk_mysql_alived
  }
  virtual_ipaddress {
    10.211.55.100 
  }
}

以下是对Keepalived配置文件的说明

vrrp_script chk_mysql_alived {        #检测pg服务是否在运行
  script "/usr/local/scripts/chk_mysql_alived.sh"
  interval 2                                        #脚本执行间隔，每2s检测一次
  weight 2
}

global_defs {
  router_id LVS_DEVEL_1
}

vrrp_instance VI_1 {
  state BACKUP                   #这里所有节点都定义为BACKUP
  interface enp0s5                #指定虚拟ip的网卡接口
  track_interface {
    enp0s5
  }
  virtual_router_id 58            #路由器标识，主从必须是一致的
  priority 200          #定义优先级，数字越大，优先级越高，在同一个vrrp_instance下，MASTER的优先级必须大于BACKUP的优先级。
  advert_int 1
  nopreempt           #不抢占模式，在优先级高的机器上设置即可，优先级低的机器可不设置
  authentication {
    auth_type PASS
    auth_pass 1111
  }
  track_script {
    chk_mysql_alived
  }
  virtual_ipaddress {
    10.211.55.100                #虚拟IP
  }
}

启动Keekpalived服务

systemctl start keepalived.service

查看Keepalived服务状态 systemctl status keepalived.service

root@node1:/usr/local/scripts# systemctl status keepalived.service
● keepalived.service - Keepalive Daemon (LVS and VRRP)
     Loaded: loaded (/lib/systemd/system/keepalived.service; enabled; vendor preset: enabled)
     Active: active (running) since Sun 2022-04-03 05:17:09 EDT; 5min ago
   Main PID: 63356 (keepalived)
      Tasks: 2 (limit: 2263)
     Memory: 1.6M
        CPU: 87ms
     CGroup: /system.slice/keepalived.service
             ├─63356 /usr/sbin/keepalived --dont-fork
             └─63357 /usr/sbin/keepalived --dont-fork

Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: Registering Kernel netlink command channel
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: Opening file '/etc/keepalived/keepalived.conf'.
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: WARNING - default user 'keepalived_script' for script execution does not exist - please create.
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: (/etc/keepalived/keepalived.conf: Line 26) (VI_1) track script chk_pg_alived not found, ignoring...
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: WARNING - script '/usr/local/scripts/chk_mysql_alived.sh' is not executable for uid:gid 0:0 - disabling.
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: SECURITY VIOLATION - scripts are being executed but script_security not enabled.
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: (VI_1) Ignoring track_interface enp0s5 since own interface
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: Registering gratuitous ARP shared channel
Apr 03 05:17:09 node1 Keepalived_vrrp[63357]: (VI_1) Entering BACKUP STATE (init)
Apr 03 05:17:12 node1 Keepalived_vrrp[63357]: (VI_1) Entering MASTER STATE
root@node1:/usr/local/scripts# 

查看虚ip是否启用 ip a

root@node1:/usr/local/scripts# ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp0s5: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:1c:42:8b:90:29 brd ff:ff:ff:ff:ff:ff
    inet 10.211.55.9/24 brd 10.211.55.255 scope global dynamic enp0s5
       valid_lft 1424sec preferred_lft 1424sec
    inet 10.211.55.100/32 scope global enp0s5
       valid_lft forever preferred_lft forever
    inet6 fdb2:2c26:f4e4:0:21c:42ff:fe8b:9029/64 scope global dynamic mngtmpaddr 
       valid_lft 2591556sec preferred_lft 604356sec
    inet6 fe80::21c:42ff:fe8b:9029/64 scope link 
       valid_lft forever preferred_lft forever

3、从库服务器上部署Keepalived

root@node2:/usr/local/scripts# cat /etc/keepalived/keepalived.conf

vrrp_script chk_mysql_alived {
  script "/usr/local/scripts/chk_mysql_alived.sh"
  interval 2 
  weight 2
}

global_defs {
  router_id LVS_DEVEL_1
}

vrrp_instance VI_1 {
  state BACKUP 
  interface enp0s5
  track_interface {
    enp0s5
  }
  virtual_router_id 58
  priority 160
  advert_int 1
  nopreempt
  authentication {
    auth_type PASS
    auth_pass 1111
  }
  track_script {
    chk_mysql_alived
  }
  virtual_ipaddress {
    10.211.55.100 
  }
}

启动Keekpalived服务

systemctl start keepalived.service

查看虚ip是否启用 ip a，从库上虚ip不会启用

---

 四、测试高可用

1、把主库的mysql服务停掉

主库停mysql服务

mysqladmin -uroot -proot shutdown

 确认主库虚ip已经消失 ip a

root@node1:/usr/local/scripts# ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp0s5: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:1c:42:8b:90:29 brd ff:ff:ff:ff:ff:ff
    inet 10.211.55.9/24 brd 10.211.55.255 scope global dynamic enp0s5
       valid_lft 994sec preferred_lft 994sec
    inet6 fdb2:2c26:f4e4:0:21c:42ff:fe8b:9029/64 scope global dynamic mngtmpaddr 
       valid_lft 2591807sec preferred_lft 604607sec
    inet6 fe80::21c:42ff:fe8b:9029/64 scope link 
       valid_lft forever preferred_lft forever

 确认主库Keepalived已经停止 systemctl status keepalived.service

root@node1:/usr/local/scripts# systemctl status keepalived.service
● keepalived.service - Keepalive Daemon (LVS and VRRP)
     Loaded: loaded (/lib/systemd/system/keepalived.service; enabled; vendor preset: enabled)
     Active: inactive (dead) since Sun 2022-04-03 05:59:18 EDT; 12s ago
    Process: 65224 ExecStart=/usr/sbin/keepalived --dont-fork $DAEMON_ARGS (code=exited, status=0/SUCCESS)
   Main PID: 65224 (code=exited, status=0/SUCCESS)
        CPU: 295ms

Apr 03 05:58:24 node1 Keepalived_vrrp[65225]: VRRP_Script(chk_mysql_alived) succeeded
Apr 03 05:58:24 node1 Keepalived_vrrp[65225]: (VI_1) Changing effective priority from 200 to 202
Apr 03 05:58:28 node1 Keepalived_vrrp[65225]: (VI_1) Entering MASTER STATE
Apr 03 05:59:17 node1 systemd[1]: Stopping Keepalive Daemon (LVS and VRRP)...
Apr 03 05:59:17 node1 Keepalived[65224]: Stopping
Apr 03 05:59:17 node1 Keepalived_vrrp[65225]: (VI_1) sent 0 priority
Apr 03 05:59:18 node1 Keepalived_vrrp[65225]: Stopped
Apr 03 05:59:18 node1 Keepalived[65224]: Stopped Keepalived v2.1.5 (07/13,2020)
Apr 03 05:59:18 node1 systemd[1]: keepalived.service: Succeeded.
Apr 03 05:59:18 node1 systemd[1]: Stopped Keepalive Daemon (LVS and VRRP).
root@node1:/usr/local/scripts# 

2、确认虚ip已经漂移到从库上  ip a

root@node2:/usr/local/scripts# ip a
1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp0s5: mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:1c:42:47:2f:1d brd ff:ff:ff:ff:ff:ff
    inet 10.211.55.4/24 brd 10.211.55.255 scope global dynamic enp0s5
       valid_lft 1386sec preferred_lft 1386sec
    inet 10.211.55.100/32 scope global enp0s5
       valid_lft forever preferred_lft forever
    inet6 fdb2:2c26:f4e4:0:21c:42ff:fe47:2f1d/64 scope global dynamic mngtmpaddr 
       valid_lft 2591966sec preferred_lft 604766sec
    inet6 fe80::21c:42ff:fe47:2f1d/64 scope link 
       valid_lft forever preferred_lft forever
root@node2:/usr/local/scripts# 

---

 五、进阶问题 

上述实验中，node1已经坏掉，node2现在是主节点，那么问题来了，如何以node2为主节点，重新搭一台从库，并在这台从库上启动Keepalived，以保证当node2坏掉的时候虚ip可以成功切换到新搭建的从库上呢？