Redis 高可用方案Keepalived

在redis3.0没有现世之前，为了提高redis的高可用性，通常的做法是采用HAProxy或Keepalived来实现故障自动转移，事实上如果是做Failover而非负载均衡的话，Keepalived的效率肯定是超过HAProxy的，所以下面我们采用Keepalived的方案。方案的架构图如下：

上图中使用多组Master-Slave构成redis集群服务，在master和slave中使用Keepalived实现故障自动转移，在客户端实现对数据的一致性哈希储存，轻松的实现水平扩展。当然在实际使用过程可以考虑一主多从的方式使用Keepalived实现故障自动转移，并实现读写分离。

如下介绍一下使用配置：

一、环境配置
master：192.168.36.54
slave：192.168.36.189
VIP：192.168.36.100
二、设计思路
当Master与Slave均运作正常时, Master负责服务，Slave负责Standby；
当Master挂掉，Slave正常时, Slave接管服务，同时关闭主从复制功能；
当Master恢复正常，则从Slave同步数据，同步数据之后关闭主从复制功能，恢复Master身份，于此同时Slave等待Master同步数据完成之后，恢复Slave身份。
然后依次循环。
需要注意的是，这样做需要在Master与Slave上都开启本地化策略，否则在互相自动切换的过程中，未开启本地化的一方会将另一方的数据清空，造成数据完全丢失。
三、使用配置
1.redis安装，参考：http://blog.csdn.net/zhu_tianwei/article/details/44890579
2.安装配置Keepalived(root)
1）安装
yum -y install keepalived
默认安装完成keepalived之后是没有配置文件的，因此我们需要手动创建。
2）配置
在主服务器上：

mkdir /etc/keepalived
vi /etc/keepalived/keepalived.conf

vrrp_script chk_redis { 
                script "/etc/keepalived/scripts/redis_check.sh"   ###监控脚本 
                interval 2                                        ###监控时间 
} 
vrrp_instance VI_1 { 
        state MASTER                            ###设置为MASTER
        interface eth0                          ###监控网卡    
        virtual_router_id 50
        priority 100                            ###权重值
        authentication { 
                     auth_type PASS             ###加密 
                     auth_pass redis            ###密码 
        } 
        track_script { 
                chk_redis                       ###执行上面定义的chk_redis
        } 
        virtual_ipaddress { 
             192.168.36.100                        ###VIP 
        }
        notify_master /etc/keepalived/scripts/redis_master.sh
        notify_backup /etc/keepalived/scripts/redis_backup.sh
        notify_fault  /etc/keepalived/scripts/redis_fault.sh
        notify_stop   /etc/keepalived/scripts/redis_stop.sh 
} 

在从服务器上：
mkdir /etc/keepalived
vi /etc/keepalived/keepalived.conf

vrrp_script chk_redis { 
                script "/etc/keepalived/scripts/redis_check.sh"   ###监控脚本 
                interval 2                                        ###监控时间 
} 
vrrp_instance VI_1 { 
        state BACKUP                                ###设置为BACKUP 
        interface eth0                              ###监控网卡
        virtual_router_id 50 
        priority 99                                ###比MASTRE权重值低 
        authentication { 
                     auth_type PASS 
                     auth_pass redis                ###密码与MASTRE相同
        } 
        track_script { 
                chk_redis                           ###执行上面定义的chk_redis
        } 
        virtual_ipaddress { 
             192.168.36.100                       ###VIP 
        } 
        notify_master /etc/keepalived/scripts/redis_master.sh
        notify_backup /etc/keepalived/scripts/redis_backup.sh
        notify_fault  /etc/keepalived/scripts/redis_fault.sh
        notify_stop   /etc/keepalived/scripts/redis_stop.sh 
}

3）创建监控脚本
在Master和Slave上创建监控Redis的脚本
$ mkdir /etc/keepalived/scripts
$ vi /etc/keepalived/scripts/redis_check.sh

#!/bin/bash

ALIVE=`/home/slim/redis/bin/redis-cli -h 192.168.36.54 -p 6379 PING`
if [ "$ALIVE" == "PONG" ]; then
  echo $ALIVE
  exit 0
else
  echo $ALIVE
  exit 1
fi

脚本中ip和端口为本机的redis服务地址，根据实际情况配置。
编写以下负责运作的关键脚本：
notify_master /etc/keepalived/scripts/redis_master.sh
notify_backup /etc/keepalived/scripts/redis_backup.sh
notify_fault /etc/keepalived/scripts/redis_fault.sh
notify_stop /etc/keepalived/scripts/redis_stop.sh
因为Keepalived在转换状态时会依照状态来呼叫：
当进入Master状态时会呼叫notify_master
当进入Backup状态时会呼叫notify_backup
当发现异常情况时进入Fault状态呼叫notify_fault
当Keepalived程序终止时则呼叫notify_stop
首先，在Redis Master上创建notity_master与notify_backup脚本：
$vi /etc/keepalived/scripts/redis_master.sh

#!/bin/bash

REDISCLI="/home/slim/redis/bin/redis-cli"
LOGFILE="/var/log/keepalived-redis-state.log"

echo "[master]" >> $LOGFILE
date >> $LOGFILE
echo "Being master...." >> $LOGFILE 2>&1

echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.36.189 6379 >> $LOGFILE  2>&1
sleep 10 #延迟10秒以后待数据同步完成后再取消同步状态

echo "Run SLAVEOF NO ONE cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF NO ONE >> $LOGFILE 2>&1

当获取到master通知时，先开启从角色状态从192.168.36.189同步数据，完成之后再关闭从角色状态。

$ vi /etc/keepalived/scripts/redis_backup.sh

#!/bin/bash

REDISCLI="/home/slim/redis/bin/redis-cli"
LOGFILE="/var/log/keepalived-redis-state.log"

echo "[backup]" >> $LOGFILE
date >> $LOGFILE
echo "Being slave...." >> $LOGFILE 2>&1

sleep 15 #延迟15秒待数据被对方同步完成之后再切换主从角色
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.36.189 6379 >> $LOGFILE  2>&1

当获取到slave通知时，等待数据同步完成，切换成从角色。
在Redis Slave上创建notity_master与notify_backup脚本：

$ vi /etc/keepalived/scripts/redis_master.sh

#!/bin/bash

REDISCLI="/home/slim/redis/bin/redis-cli"
LOGFILE="/var/log/keepalived-redis-state.log"

echo "[master]" >> $LOGFILE
date >> $LOGFILE
echo "Being master...." >> $LOGFILE 2>&1

echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.36.54 6379 >> $LOGFILE  2>&1
sleep 10 #延迟10秒以后待数据同步完成后再取消同步状态

echo "Run SLAVEOF NO ONE cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF NO ONE >> $LOGFILE 2>&1

$ vi /etc/keepalived/scripts/redis_backup.sh

#!/bin/bash

REDISCLI="/home/slim/redis/bin/redis-cli"
LOGFILE="/var/log/keepalived-redis-state.log"

echo "[backup]" >> $LOGFILE
date >> $LOGFILE
echo "Being slave...." >> $LOGFILE 2>&1

sleep 15 #延迟15秒待数据被对方同步完成之后再切换主从角色
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.36.54 6379 >> $LOGFILE  2>&1

然后在Master与Slave创建如下相同的脚本：

$ vi /etc/keepalived/scripts/redis_fault.sh

#!/bin/bash

LOGFILE=/var/log/keepalived-redis-state.log

echo "[fault]" >> $LOGFILE
date >> $LOGFILE

$ vi /etc/keepalived/scripts/redis_stop.sh

#!/bin/bash

LOGFILE=/var/log/keepalived-redis-state.log

echo "[stop]" >> $LOGFILE
date >> $LOGFILE

给脚本都加上可执行权限：
$ chmod +x /etc/keepalived/scripts/*.sh

3.启动服务

脚本创建完成以后，我们开始按照如下流程进行测试：
1)启动Master上的Redis(slim)
$./bin/redis-server conf/redis.conf ;tail -f logs/redis.log
2)启动Slave上的Redis(slim)
$./bin/redis-server conf/redis.conf ;tail -f logs/redis.log
3)启动Master上的Keepalived(root)
$/etc/init.d/keepalived start
4)启动Slave上的Keepalived(root)
$/etc/init.d/keepalived start

4.测试

1)尝试通过VIP连接Redis:
$ ./bin/redis-cli -h 192.168.36.100 -p 6379 INFO
连接成功，Slave也连接上来了。
role:master
slave0:192.168.36.189,6379,online
2)尝试插入一些数据：
$ ./bin/redis-cli -h 192.168.36.100 -p 6379 SET Hello Redis
OK
从VIP读取数据
$ ./bin/redis-cli -h 10.6.1.200 GET Hello
“Redis”
从Master读取数据
$ ./bin/redis-cli -h 192.168.36.54 -p 6379 GET Hello
“Redis”
从Slave读取数据
$ ./bin/redis-cli -h 192.168.36.189 -p 6379 GET Hello
“Redis”
4.模拟故障
将Master上的Redis进程杀死：
$killall -9 redis-server
查看Master上的Keepalived日志
$ tailf /var/log/keepalived-redis-state.log
[fault]

同时Slave上的日志显示：
$ tailf /var/log/keepalived-redis-state.log
[master]
Being master….
Run SLAVEOF cmd …
OK
Run SLAVEOF NO ONE cmd …
OK

然后我们可以发现，Slave已经接管服务，并且担任Master的角色了。
$ redis-cli -h 192.168.36.100 -p 6379 INFO
$ redis-cli -h 192.168.36.189 -p 6379 INFO
role:master
然后我们恢复Master的Redis进程(slim)
$ ./bin/redis-server conf/redis.conf ;tail -f logs/redis.log
查看Master上的Keepalived日志
$ tailf /var/log/keepalived-redis-state.log
[master]
Being master….
Run SLAVEOF cmd …
OK
Run SLAVEOF NO ONE cmd …
OK
同时Slave上的日志显示：
$ tailf /var/log/keepalived-redis-state.log
[backup]
Being slave….
Run SLAVEOF cmd …
OK
可以发现目前的Master已经再次恢复了Master的角色，故障切换以及自动恢复都成功了。大家可以测试一下再故障情况下添加数据，服务正常后，数据是否正常同步。

文章参考： http://heylinux.com/archives/1942.html