redis主备部署方案

Posted Novicelong

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了redis主备部署方案相关的知识,希望对你有一定的参考价值。

Redis部署方式采用主备的方式,通过keepalived来对外提供虚IP,并实现主备自动切换功能。

主实例A:192.168.20.30

备实例B:192.168.20.232

虚IP:192.168.20.110

正常工作时,虚IP在主实例A上,主实例A上的数据自动同步到备实例B上,当主实例A挂掉之后,备实例B将自动接管虚IP,并将redis转换为主模式,待原主实例A恢复后,A将自动切换成备模式,从B上同步数据,主备角色互换,实现融灾备份。

 

安装部署步骤如下:

  1. 1.       安装keepalived

wget http://www.keepalived.org/software/keepalived-1.2.6.tar.gz

tar zxvf keepalived-1.2.6.tar.gz

cd keepalived-1.2.6

./configure

如果报错

configure: error:

        !!! OpenSSL is not properly installed on your system. !!!

  !!! Can not include OpenSSL headers files.

解决办法:

yum -y install openssl-devel

yum -y install popt-devel

ln -s /usr/src/kernels/2.6.32-220.el6.x86_64/ /usr/src/linux

 

 

./configure

make

make install

 

cp /usr/local/etc/rc.d/init.d/keepalived /etc/rc.d/init.d/

cp /usr/local/etc/sysconfig/keepalived /etc/sysconfig/

cp /usr/local/sbin/keepalived /usr/sbin/

mkdir /etc/keepalived

添加keepalived的配置项:

vi /etc/keepalived/keepalived.conf

global_defs {

   router_id LVS_DEVEL

}

vrrp_script Monitor_Redis {

 script "/home/bbcv/redis/redis_keepalive.sh"

 interval 3 #每3秒执行一次

 weight 2

}

# VIP1

vrrp_instance VI_1 {

    state BACKUP           #主备服务器都设置成BACKUP

    interface eth0

    virtual_router_id 55

    priority 100     # 备份服务上将100改为90

    advert_int 1    #检查间隔

    nopreempt     #设置为不抢占,注意这个配置只能设置在state为BACKUP的主机上,而且这个主机的priority必须比另外一台高

    authentication {

        auth_type PASS

        auth_pass 1111

    }

    notify_master /home/bbcv/redis/redis_master.sh

    notify_backup /home/bbcv/redis/redis_backup.sh

    track_script {

        Monitor_Redis #(调用redis进程检测脚本)

    }

    virtual_ipaddress {

        192.168.20.110

        #(如果有多个VIP,继续换行填写.)

    }

}

启动keepalived

service keepalived start

  1. 2.       redis_keepalive.sh脚本

该脚本主要实现对redis进程进行监控,当检测到redis进程挂掉时,自动停止keepalived进程,使虚IP进行漂移

²  注意添加该脚本的可执行权限!

chmod +x redis_keepalive.sh

#!/bin/bash

time=$(date \'+%Y-%m-%d %H:%M:%S\' )

#redis部署路径

redispath=/home/bbcv/redis

#redis-cli部署路径

rediscli=$redispath/redis-2.0.0-rc4/redis-cli

if [ ! -d "logs" ]; then

        mkdir $redispath/logs

fi

logfile=$redispath/logs/redis-state.log

oldfile=$logfile\'.\'$(date +%Y-%m-%d --date=\'30 days ago\')

yesterdayfile=$logfile\'.\'$(date +%Y-%m-%d --date=\'1 days ago\')

#把昨天的日志重命名

if [ -f $yesterdayfile ]

then

        echo "OK"

else

        mv $logfile $yesterdayfile >> $logfile 2>&1

        echo "$time [$yesterdayfile] Move yesterdayfile Success!" >> $logfile

fi

#删除老的日志文件

if [ -f $oldfile ]

then

        rm -f $oldfile >> $logfile 2>&1

        echo "$time [$oldfile] Delete Old File Success!" >> $logfile

else

        echo "no old file"

fi

#检测redis端口是否正常

status=`$rediscli -h 127.0.0.1 -p 6379 info|grep role|awk -F ":" \'{print $2}\'`

$rediscli -h 127.0.0.1 -p 6379 info > /dev/null

if [ $? -eq 0 ]; then

        echo "$time redis server is OK" >>$logfile

        echo 目前状态为:$status >>$logfile

else

        echo "$time no redis service found!" >>$logfile

        sleep 2

        # try to found it again

        $rediscli -h 127.0.0.1 -p 6379 info > /dev/null

        if [ $? -eq 0 ]; then

                echo "$time redis server is OK" >>$logfile

                echo 目前状态为:$status >>$logfile

                #exit 0

        else

                echo "$time redis server error" >>$logfile

                #stop keepalived

                /etc/init.d/keepalived stop

                echo "$time stop keepalived" >>$logfile

        fi

fi

  1. 3.       redis_master.sh脚本

该脚本主要实现当keepalived切换到master时,执行slaveof no one语句把redis的端口转换为主模式

²  注意添加该脚本的可执行权限!

chmod +x redis_master.sh

#!/bin/sh

#本机IP地址

localip=127.0.0.1

#另一台服务器的地址

backip=192.168.20.232

#redis端口号

port="6379 9998 11111 14948"

 

time=$(date \'+%Y-%m-%d %H:%M:%S\' )

#redis部署路径

redispath=/home/bbcv/redis

#redis-cli部署路径

rediscli=$redispath/redis-2.0.0-rc4/redis-cli

#日志文件目录

logfile=$redispath/logs/redis-state.log

for p in $port

do

{

        echo "$time redis $p端口切换成主端口" >>$logfile

        $rediscli -h $localip -p $p slaveof no one >>$logfile 2>&1

        sleep 1

}

done

  1. 4.       redis_backup.sh

该脚本主要实现当keepalived切换到master时,执行slaveof 主redisIP 端口 语句,把redis的端口转换为备模式,并同步主redis的数据

²  注意添加该脚本的可执行权限!

chmod +x redis_backup.sh

#!/bin/sh

#本机IP地址

localip=127.0.0.1

#另一台服务器的地址

backip=192.168.20.232

#redis端口号

port="6379 9998 11111 14948"

 

time=$(date \'+%Y-%m-%d %H:%M:%S\' )

#redis部署路径

redispath=/home/bbcv/redis

#redis-cli部署路径

rediscli=$redispath/redis-2.0.0-rc4/redis-cli

#日志文件目录

logfile=$redispath/logs/redis-state.log

for p in $port

do

{

        echo "$time redis $p端口切换成$backip的备端口" >>$logfile

        $rediscli -h $localip -p $p slaveof $backip $p  >>$logfile 2>&1

        sleep 1

}

done

  1. 5.       安装redis

1.新建文件夹,如mkdir redis

2.将redis-2.2.13.tar.gz拷入其中

3.解压并进入主目录

4.make

5.make install(Ubuntu上需要sudo执行)

6.修改配置文件redis.conf

修改如下:

daemonize no 改为 daemonize yes

取消注释:

syslog-enabled no (改为syslog-enabled yes )

syslog-facility local0

已将修改后的redis.conf文件放入附件,不同需求还需再修改配置文件(如修改端口等)

7.在执行redis-server redis.conf启动时,可能会报如下错误:

Can\'t chdir to \'/var/db/redis\': No such file or directory

需要手动创建\'/var/db/redis\'文件夹

# WARNING overcommit_memory is set to 0! Background save may fail under low memory condition

需要执行 sysctl vm.overcommit_memory=1 修改内核参数

8.再次执行redis-server redis.conf可成功启动redis

  1. 6.       测试

1.查看当前redis是主还是备

/home/bbcv/redis/redis-2.0.0-rc4/redis-cli -h 127.0.0.1 -p 6379 info

 

2.在主redisA上新增一条数据,查看备redisB是否同步该数据

/home/bbcv/redis/redis-2.0.0-rc4/redis-cli -h 127.0.0.1 -p 6379

 

3.停止主redisA,查看虚IP是否切换到备redisB服务器,并查看备redisB服务器是否切换成主状态

ip a查看虚IP是否在本机上

 

4.恢复redisA,查看redisA是否从redisB中同步数据

 

百度原文

 http://wenku.baidu.com/view/511754254b73f242336c5f72.html

以上是关于redis主备部署方案的主要内容,如果未能解决你的问题,请参考以下文章

Oracle Dataguard HA (主备,灾备)方案部署调试

使用Redis分布式锁实现主备

使用Redis分布式锁实现主备

redis演练 redis复制(主备模式)

搭建和测试 Redis 主备和集群

keepalived实现redis主备切换