mysql 高性能搭建2-1： Centos7离线按照keepalived

一 Keepalived概述

1.1 keepalived概述

 keepalived是vrrp协议的实现，原生设计目的是为了高可用ipvs服务，keepalived能够配置文件中定义生成ipvs规则，并能够对各RS的健康状态进行检测；通过共用的虚拟IP地址对外提供服务；每个热备组内同一时刻只有一台主服务器提供服务，其他服务器处于冗余状态，若当前在线的服务器宕机，其虚拟IP地址将会被其他服务器接替（优先级决定接替顺序），实现高可用为后端主机提供服务。

1.2 vrrp概述

VRRP是一种选择协议，它可以把一个虚拟路由器的责任动态分配到局域网上的VRRP路由器中的一台。

控制虚拟路由器IP地址的VRRP路由器称为主路由器，它负责转发数据包到这些虚拟IP地址。一旦主路由器不可用，这种选择过程就提供了动态的故障转移机制，这就允许虚拟路由器的IP地址可以作为终端主机的默认第一跳路由器。

1.3 keepalived的模式

keepalive通过组播，单播等方式(自定义)，实现keepalive主备推选。工作模式分为抢占和非抢占（通过参数nopreempt来控制）。

1）抢占模式：

主服务正常工作时，虚拟IP会在主上，备不提供服务，当主服务优先级低于备的时候，备会自动抢占虚拟IP，这时，主不提供服务，备提供服务。

也就是说，工作在抢占模式下，不分主备，只管优先级。

如上配置，不管keepalived.conf里的state配置成master还是backup，只看谁的priority优先级高（一般而言，state为MASTER的优先级要高于BACKUP）。priority优先级高的那一个在故障恢复后，会自动将VIP资源再次抢占回来！！

2）非抢占模式：

这种方式通过参数nopreempt（一般设置在advert_int的那一行下面）来控制。不管priority优先级，只要MASTER机器发生故障，VIP资源就会被切换到BACKUP上。并且当MASTER机器恢复后，也不会去将VIP资源抢占回来，直至BACKUP机器发生故障时，才能自动切换回来。

千万注意：

nopreempt这个参数只能用于state为backup的情况，所以在配置的时候要把master和backup的state都设置成backup，这样才会实现keepalived的非抢占模式！

也就是说：

A)当state状态一个为master，一个为backup的时候，加不加nopreempt这个参数都是一样的效果。即都是根据priority优先级来决定谁抢占vip资源的，是抢占模式！

B)当state状态都设置成backup，如果不配置nopreempt参数，那么也是看priority优先级决定谁抢占vip资源，即也是抢占模式。

C)当state状态都设置成backup，如果配置nopreempt参数，那么就不会去考虑priority优先级了，是非抢占模式！即只有vip当前所在机器发生故障，另一台机器才能接管vip。即使优先级高的那一台机器恢复后也不会主动抢回vip，只能等到对方发生故障，才会将vip切回来。

1.4 keepalived的组件

core:keepalived核心组件，主进程的启动和维护，全局配置等。

vrrp stack:keepalived是基于vrrp协议实现高可用vps服务，vrrp则为相关子进程为其提供服务 

check:检测keepalived的健康状态相关进程 

system call:系统调用

watch dog:监控check和vrrp进程的看管者，check负责检测器子进程的健康状态，当其检测到master上的服务不可用时则通告vrrp将其转移至backup服务器上。

二  安装keepalived的依赖

2.1 安装keepalived的依赖软件包

https://blog.youkuaiyun.com/Loiterer_Y/article/details/104468964/

软件包位置：见百度网盘中软件包

离线安装keepalived的所有依赖，如下图所示：

 下载所有依赖之后，将其上传到服务器,然后在依赖所在文件夹内执行

rpm -Uvh --force *.rpm即可。

2.2 安装keepalived

2.2.1 安装keepalived的版本

2.2.2 进行安装

进入到/usr/local/src目录下：

[root@bogon src]# pwd /usr/local/src [root@bogon src]# tar -zxvf /root/bigdata-software/keepalived-1.4.0.tar.gz  -C . keepalived-1.4.0/ keepalived-1.4.0/install-sh keepalived-1.4.0/missing keepalived-1.4.0/compile [root@bogon src]# cd   keepalived-1.4.0 [root@bogon keepalived-1.4.0]# ls aclocal.m4  AUTHOR       ChangeLog  configure     CONTRIBUTORS  depcomp  genhash  install-sh  keepalived.spec.in  Makefile.am  missing    snap ar-lib      bin_install  compile    configure.ac  COPYING       doc      INSTALL  keepalived  lib                 Makefile.in  README.md  TODO [root@bogon keepalived-1.4.0]# ./configure --prefix=/usr/local/keepalived checking for a BSD-compatible install... /usr/bin/install -c [root@bogon keepalived-1.4.0]# make && make install Making all in lib make[1]: Entering directory `/usr/local/src/keepalived-1.4.0/lib' [root@bogon keepalived-1.4.0]# cp /usr/local/src/keepalived-1.4.0/keepalived/etc/init.d/keepalived /etc/rc.d/init.d/ cp: overwrite ‘/etc/rc.d/init.d/keepalived’? y [root@bogon keepalived-1.4.0]# cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/ cp: overwrite ‘/etc/sysconfig/keepalived’? y [root@bogon keepalived-1.4.0]# mkdir /etc/keepalived/ [root@bogon keepalived-1.4.0]#  cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/ [root@bogon keepalived-1.4.0]# cp /usr/local/keepalived/sbin/keepalived /usr/sbin/ [root@bogon keepalived-1.4.0]# echo "/etc/init.d/keepalived start" >> /etc/rc.local

2.2.3 修改配置文件

[root@bogon keepalived-1.4.0]# cd /etc/keepalived [root@bogon keepalived]# ls keepalived.conf [root@bogon keepalived]#  cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak [root@bogon keepalived]# ls keepalived.conf  keepalived.conf.bak [root@bogon keepalived]# vim /etc/keepalived/keepalived.conf

修改内容：

! Configuration File for keepalived global_defs { notification_email { ops@wangshibo.cn tech@wangshibo.cn } notification_email_from ops@wangshibo.cn smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id MASTER-HA } vrrp_script chk_mysql_port {     #检测mysql服务是否在运行。有很多方式，比如进程，用脚本检测等等     script "/opt/chk_mysql.sh"   #这里通过脚本监测     interval 2                   #脚本执行间隔，每2s检测一次     weight -5                    #脚本结果导致的优先级变更，检测失败（脚本返回非0）则优先级 -5     fall 2                    #检测连续2次失败才算确定是真失败。会用weight减少优先级（1-255之间）     rise 1                    #检测1次成功就算成功。但不修改优先级 } vrrp_instance VI_1 {     state MASTER     interface eth0      #指定虚拟ip的网卡接口     mcast_src_ip 192.168.152.136     virtual_router_id 51    #路由器标识，MASTER和BACKUP必须是一致的     priority 101            #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。这样MASTER故障恢复后，就可以将VIP资源再次抢回来     advert_int 1     authentication {         auth_type PASS         auth_pass 1111     }     virtual_ipaddress {         192.168.152.130     } track_script {    chk_mysql_port } }

需要注意的地方：

2.2.4 脚本配置

[root@bogon keepalived]# vim /opt/chk_mysql.sh

修改内容：

#!/bin/bash port=`netstat -lntup |grep 3306|wc -l` echo $port if [ $port -ne 1 ] then  /etc/init.d/keepalived stop else  echo "Mysql is running" fi

赋予执行权限

[root@bogon keepalived]# chmod 755 /opt/chk_mysql.sh

2.2.5 启动keeepalived

[root@bogon keepalived]#  /etc/init.d/keepalived start Starting keepalived (via systemctl):                       [  OK  ] [root@bogon keepalived]#  /etc/init.d/keepalived status ● keepalived.service - LVS and VRRP High Availability Monitor    Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)    Active: active (running) since Mon 2022-07-25 08:27:09 CST; 4s ago   Process: 24336 ExecStart=/usr/local/keepalived/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)  Main PID: 24337 (keepalived)     Tasks: 1    CGroup: /system.slice/keepalived.service            └─24337 /usr/local/keepalived/sbin/keepalived -D Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Registering Kernel netlink reflector Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Registering Kernel netlink command channel Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Registering gratuitous ARP shared channel Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Opening file '/etc/keepalived/keepalived.conf'. Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: WARNING - default user 'keepalived_script' for script execution does not exist - please create. Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Cant find interface eth0 for vrrp_instance VI_1 !!! Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: Default interface eth0 does not exist and no interface specified. Skipping static address 192.168.152.130. Jul 25 08:27:09 bogon Keepalived_vrrp[24339]: VRRP_Instance(VI_1) Unknown interface ! Jul 25 08:27:10 bogon Keepalived[24337]: Keepalived_vrrp exited with permanent error CONFIG. Terminating Jul 25 08:27:10 bogon Keepalived[24337]: Stopping

效果截图：

2.2.6 另一种mysql检测脚本

#!/bin/bash  
set -x
MYSQL=/usr/bin/mysql
MYSQL_HOST=localhost 
MYSQL_USER='check_mysql'
MYSQL_PASSWORD=''
CHECK_TIME=3
#mysql  is working MYSQL_OK is 1 , mysql down MYSQL_OK is 0  
MYSQL_OK=1 
function check_mysql_health (){  
	running_bak=`ps -AF | grep "single-transaction --all-databases --triggers --routines --flush-logs" | grep -v "grep" | wc -l`;
        if [ $running_bak -ne 0 ]; then
		return $MYSQL_OK;
        fi
 
	$MYSQL -h$MYSQL_HOST -u$MYSQL_USER -e"show status;" >/dev/null 2>&1  
	if [ $? = 0 ] ;then  
		 MYSQL_OK=1 
	else  
		 MYSQL_OK=0 
	fi  
		 return $MYSQL_OK  
}  
 
while [ $CHECK_TIME -ne 0 ]  
do  
     let "CHECK_TIME -= 1"  
     check_mysql_health  
     if [ $MYSQL_OK = 1 ] ; then  
          CHECK_TIME=0 
          exit 0  
     fi  
 
     if [ $MYSQL_OK -eq 0 ] &&  [ $CHECK_TIME -eq 0 ]  
     then  
        /etc/init.d/keepalived stop  
     	exit 1   
     fi  
     sleep 1  
done

 check_mysql.sh_疯飙的蜗牛的博客-优快云博客

三  在线安装keepalived的依赖（了解）

yum   install    keepalived   -y