Redis5主备安装-Redis

本次Redis有两台服务器及3个独立IP：主服务器的ip地址是192.168.31.190，从服务器的IP地址是192.168.31.191，vipIP地址是192.168.31.216

主备方案承载Redis最大的好处是无需考虑Redis崩后无法访问。

 前提是需要优先安装keepalived，其次安装Redis。

1、master配置

! Configuration File for keepalivedglobal_defs {router_id master
}vrrp_script chk_redis {script "/data/keepalived/redis_check.sh"interval 2    
}vrrp_instance VI_1 {state MASTERinterface ens33virtual_router_id 51priority 150advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.31.216 dev ens33 label ens33:1}track_script {chk_redis}notify_master "/data/keepalived/redis_master.sh master"notify_backup /data/keepalived/redis_backup.shnotify_fault /data/keepalived/redis_fault.shnotify_stop /data/keepalived/redis_stop.sh
}

2、backup1配置

! Configuration File for keepalived
global_defs {router_id backup1
}
vrrp_script chk_redis {script "/data/keepalived/redis_check.sh"interval 2
}
vrrp_instance VI_1 {state BACKUPinterface ens33virtual_router_id 51priority 80advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.31.216 dev ens33 label ens33:1}track_script {chk_redis  }notify_master "/data/keepalived/redis_master.sh master"notify_backup /data/keepalived/redis_backup.shnotify_fault /data/keepalived/redis_fault.shnotify_stop /data/keepalived/redis_stop.sh
}

3、创建脚本目录

mkdir -p /data/keepalived

4、编辑监听脚本

vi /data/keepalived/redis_check.sh
#!/bin/bash
ALIVE=`redis-cli -a 123456 PING` 
if [ "$ALIVE" == "PONG" ];thenecho $ALIVEexit 0
elsesystemctl stop keepalived.serviceexit 1
fi

5、默认监听脚本

vi /data/keepalived/redis_fault.sh
#!/bin/bash 
LOGFILE=/data/keepalived/redis-state.log
echo "[fault]" >> $LOGFILE 
date >> $LOGFILE

6、停止监听脚本

vi /data/keepalived/redis_stop.sh
#!/bin/bash 
LOGFILE=/data/keepalived/redis-state.log
echo "[stop]" >> $LOGFILE 
date >> $LOGFILE

7、master服务器master脚本

vi /data/keepalived/redis_master.sh
#!/bin/bash
REDISCLI="redis-cli -a 123456"
LOGFILE="/data/keepalived/redis-state.log"
echo "[master]" >> $LOGFILE
date >> $LOGFILE
echo "Being master...." >> $LOGFILE 2>&1
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.31.191 6379 >> $LOGFILE  2>&1
sleep 10
echo "Run REPLICAOF NO ONE cmd ..." >> $LOGFILE
$REDISCLI REPLICAOF NO ONE >> $LOGFILE 2>&1
date >> $LOGFILE
在该节点的keepalived第一次启动，指keepalived组中仅有这一台或者这一台的优先级别最高，自动成为主节点，然后就会调用该脚本；
脚本中先从备机上同步数据，如果备机没有运行或故障了，则过几秒执行命令使这个节点的Redis成为主节点（提供读写能力）。
邮件配置
contact=([e-mail]) #接收者邮箱，多个以空格分隔
HOST_IP=[192.168.31.190] #本机ip
notify() {mailsubject="Redis发生切换($HOST_IP)" #邮件主题mailbody="$(date +'%F %T'): vrrp transition, $HOST_IP keepalived changed to be $1" #邮件正文for receiver in ${contact[*]}doecho "$mailbody" | mail -s "$mailsubject" $receiver #发送邮件done
}
case $1 in
master)notify master;;
backup)notify backup;;
fault)notify fault;;
*)echo "Usage:{master|backup|fault}"exit 1;;
esac

8、master服务器backup脚本

vi /data/keepalived/redis_backup.sh
#!/bin/bash
REDISCLI="redis-cli -a 123456"
LOGFILE="/data/keepalived/redis-state.log"
echo "[backup]" >> $LOGFILE
date >> $LOGFILE
echo "Being slave...." >> $LOGFILE 2>&1
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.31.191 6379 >> $LOGFILE  2>&1
sleep 15
echo "Run REPLICAOF NO ONE cmd ..." >> $LOGFILE
$REDISCLI REPLICAOF NO ONE >> $LOGFILE 2>&1
当这个原来是主节点，但是中途挂了，最后又重新启动后，这个时候已经由一个备节点变成了主节点，则keepalived会执行该脚本；
该脚本首先等一段时间，给备机充足的时间切换为主节点，然后该脚本执行命令，让本节点的Redis成为新主节点Redis的备份服务。

9、backup1服务器master脚本

vi /data/keepalived/redis_master.sh
#!/bin/bash
REDISCLI="redis-cli -a 123456"
LOGFILE="/data/keepalived/redis-state.log"
echo "[master]" >> $LOGFILE
date >> $LOGFILE
echo "Being master...." >> $LOGFILE 2>&1
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.31.190 6379 >> $LOGFILE  2>&1
sleep 10
echo "Run SLAVEOF NO ONE cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF NO ONE >> $LOGFILE 2>&1
当该节点在主节点挂掉之后会执行；
脚本会先从主节点同步数据，然后脚本执行命令让本机Redis独立为主服务（提供读写能力）
邮件配置
contact=([e-mail]) #接收者邮箱，多个以空格分隔
HOST_IP=[192.168.31.191] #本机ip
notify() {mailsubject="Redis发生切换($HOST_IP)" #邮件主题mailbody="$(date +'%F %T'): vrrp transition, $HOST_IP keepalived changed to be $1" #邮件正文for receiver in ${contact[*]}doecho "$mailbody" | mail -s "$mailsubject" $receiver #发送邮件done
}
case $1 in
master)notify master;;
backup)notify backup;;
fault)notify fault;;
*)echo "Usage:{master|backup|fault}"exit 1;;
esac

10、backup1服务器backup脚本

vi /data/keepalived/redis_backup.sh
#!/bin/bash
REDISCLI="redis-cli -a 123456"
LOGFILE="/data/keepalived/redis-state.log"
echo "[backup]" >> $LOGFILE
date >> $LOGFILE
echo "Being slave...." >> $LOGFILE 2>&1
sleep 15
echo "Run SLAVEOF cmd ..." >> $LOGFILE
$REDISCLI SLAVEOF 192.168.31.190 6379 >> $LOGFILE  2>&1
该脚本是在主节点的Redis成为主服务时，或主节点服务已经开启，备机服务再开启后自动执行；
脚本先等待一段时间，让主节点切换为主服务，然后脚本执行命令使本节点的Redis成为主节点的备机。

11、授权

cd /data/keepalived/
chmod +x *.sh
touch redis-state.log

12、启动注意事项

#优先启动master服务器的keepalived，但是前提是Redis已经启动完毕咯，第二是keepalived端口已经放出来咯
#第一次启动时，backup1节点的服务一定是在主master节点服务都开启后才启动，本机Redis只能提供读的功能（可配置为可写，但不安全）。
systemctl start keepalived.service #启动服务
systemctl stop keepalived.service #停止服务
systemctl restart keepalived.service #重启服务
ps -ef|grep keepalive #检查keepalive服务
ip addr #检查vip是否启动

13、测试

redis-cli -h 192.168.31.216 -p 6379
auth 123456
set id 1
get id
redis-cli -h 192.168.31.190 -p 6379
auth 123456
get id
redis-cli -h 192.168.31.191 -p 6379
auth 123456
get id
确认都有数据的情况下，停止192.168.31.190的Redis，正常来说，会将192.168.31.216IP漂移到192.168.31.191这台服务器上，然后测试
190这台服务器上的操作
systemctl stop redis.service
ps -ef|grep keepalive #检查keepalive服务
ip addr #检查vip是否启动
191这台服务器上的操作
ps -ef|grep keepalive #检查keepalive服务
ip addr #检查vip是否启动
redis-cli -h 192.168.31.216 -p 6379
auth 123456
set name wangyulong
get name
redis-cli -h 192.168.31.191 -p 6379
auth 123456
get name
再次整个漂移是成功的
如果再次启动192.168.31.190的Redis服务以及keepalive服务，正常来说又会回到以前
systemctl start redis.service #启动redis服务 
systemctl start keepalived.service #启动服务
然后到192.168.31.191这台服务器上去查看keepalive服务以及vip是否存在
ps -ef|grep keepalive #检查keepalive服务
ip addr #检查vip是否启动
正常来说，在192.168.31.191这台服务器上keepalive服务是正常运行的，而VIP却是不存在的。因为vip又回到了主服务(192.168.31.190)上
再到192.168.31.190这台服务器上去查看keepalive服务以及vip是否存在
ps -ef|grep keepalive #检查keepalive服务
ip addr #检查vip是否启动
正常来说，在192.168.31.190这台服务器上keepalive服务是正常运行的，VIP服务也是正常运行的。最后首先采用vip连接到redis服务里面，查看name是否存在
redis-cli -h 192.168.31.216 -p 6379
auth 123456
get name
redis-cli -h 192.168.31.190 -p 6379
auth 123456
get name

 前提是需要优先安装keepalived，其次安装Redis。