DoubleLi

qq: 517712484 wx: ldbgliet
  博客园 :: 首页 :: 博问 :: 闪存 :: 新随笔 :: 联系 :: 订阅 订阅 :: 管理 ::
Redis+Keepalived高可用环境部署记录

Keepalived 实现VRRP（虚拟路由冗余）协议，从路由级别实现VIP切换，可以完全避免类似heartbeat脑裂问题，可以很好的实现主从、主备、互备方案，尤其是无状态业务，有状态业务就需要额外花些功夫了。既然Mysql可以使用Keepalived很好的做到主从切换，那么Redis自然可以使用这种方式实现高可用。

Redis主从实现完全没有Mysql成熟，仅仅是可用而已，经过测试主从也不是那么完全不靠谱，主要问题在于同步连接断开之后需要重新全量同步，如果频繁进行会对主服务带来很大性能影响。 但现实中主从机器往往要求放在一个机柜同一台交换设备下，网络闪断情况极低；再者主从同步在同步数量量大情况下，需要将缓存区调得足够大，不然也容易造成连接断开。
实现切换逻辑如下：A和B两台机器
1）A 、B机器依次启动，A机作为主、B机为从。
2）主A机挂掉，B机接管业务并作为主。
3）A机起来，作为从SLAVEOF B。
4）B机挂掉，A机再切回主。

在Keepalived 有两个角色：Master(一个)、Backup（多个），如果设置一个为Master，但Master挂了后再起来，必然再次业务又一次切换，这对于有状态服务是不可接受的。解决方案就是两台机器都设置为Backup，而且优先级高的Backup设置为nopreemt 不抢占。

部署记录：

0）服务器信息

1
2
3
4
5
6
7
8
9
10
11
12
192.168.10.205    redis-master     需要安装redis(3.2.0版本)、keepalived(1.3.2版本)
192.168.10.206    redis-slave      需要安装redis(3.2.0版本)、keepalived(1.3.2版本)
192.168.10.230    VIP
 
关闭两个节点机器的iptables和selinux（两个节点上都要操作）
[root@redis-master ~]# /etc/init.d/iptables stop
[root@redis-master ~]# vim /etc/sysconfig/selinux
......
SELINUX=disabled
[root@redis-master ~]# setenforce 0
[root@redis-master ~]# getenforce
Permissive

1）安装redis服务及主从配置（两个节点机都要操作）

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
[root@redis-master ~]# cd /usr/local/src/
[root@redis-master src]# wget http://download.redis.io/releases/redis-3.2.0.tar.gz
[root@redis-master src]# tar -zvxf redis-3.2.0.tar.gz
[root@redis-master src]# cd redis-3.2.0
[root@redis-master redis-3.2.0]# make
 
添加相关文件及命令
[root@redis-master redis-3.2.0]# mkdir -p /usr/local/redis/bin/
[root@redis-master redis-3.2.0]# cd src
[root@redis-master src]# cp redis-benchmark redis-check-aof redis-check-rdb redis-cli redis-server redis-sentinel /usr/local/redis/bin/
[root@redis-master src]# cd ../
[root@redis-master redis-3.2.0]# cp redis.conf /etc/
 
添加redis启动脚本
[root@redis-master redis-3.2.0]# vim /etc/init.d/redis
#!/bin/bash
#chkconfig: 2345 10 90
#description: Start and Stop redis
 
REDISPORT=6379
 
EXEC=/usr/local/redis/bin/redis-server
 
REDIS_CLI=/usr/local/redis/bin/redis-cli
 
PIDFILE=/var/run/redis.pid
 
CONF="/etc/redis.conf"
 
case "$1" in
 
start)
 
if [ -f $PIDFILE ]
 
then
 
echo "$PIDFILE exists, process is already running or crashed"
 
else
 
echo "Starting Redis server..."
 
$EXEC $CONF
 
fi
 
if [ "$?"="0" ]
 
then
 
echo "Redis is running..."
 
fi
 
;;
 
stop)
 
if [ ! -f $PIDFILE ]
 
then
 
echo "$PIDFILE does not exist, process is not running"
 
else
 
PID=$(cat $PIDFILE)
 
echo "Stopping ..."
 
$REDIS_CLI -p $REDISPORT SHUTDOWN
 
while [ -x ${PIDFILE} ]
 
do
 
echo "Waiting for Redis to shutdown ..."
 
sleep 1
 
done
 
echo "Redis stopped"
 
fi
 
;;
 
restart|force-reload)
 
${0} stop
 
${0} start
 
;;
 
*)
 
echo "Usage: /etc/init.d/redis {start|stop|restart|force-reload}" >&2
 
exit 1
 
esac
 
添加执行权限
[root@redis-master redis-3.2.0]# chmod 755 /etc/init.d/redis
 
设置开机自启动
[root@redis-master redis-3.2.0]# chkconfig --add redis
[root@redis-master redis-3.2.0]# chkconfig redis on
 
创建redis状态日志
[root@redis-master redis-3.2.0]# mkdir /var/log/redis/
[root@redis-master redis-3.2.0]# touch /var/log/redis/redis.log
 
redis主从配置（先看下redis-master主节点的配置）
[root@redis-master redis-3.2.0]# vim /etc/redis.conf
.......
port 6379
.......
daemonize yes              #这个修改为yes
.......
bind 0.0.0.0               #绑定的主机地址。说明只能通过这个ip地址连接本机的redis。最好绑定0.0.0.0；注意这个不能配置成127.0.0.1，否则复制会失败！用0.0.0.0或者本机ip地址都可以
.......
pidfile /var/run/redis.pid
.......
logfile /var/log/redis/redis.log
.......
dir  /var/redis/redis      #redis数据目录
.......
appendonly yes              #启用AOF持久化方式
appendfilename "appendonly.aof"       #AOF文件的名称，默认为appendonly.aof
appendfsync everysec        #每秒钟强制写入磁盘一次，在性能和持久化方面做了很好的折中，是受推荐的方式。
.....
save 900 1                 #启用RDB快照功能，默认就是启用的
save 300 10
save 60 10000              ＃即在多少秒的时间内，有多少key被改变的数据添加到.rdb文件里
.......
slave-serve-stale-data yes    #默认就会开启
slave-read-only yes
......
dbfilename dump.rdb       ＃快照文件名称
......
 
另一个从节点redis-slave的redis.conf配置和上面基本差不多，只是多了下面一行配置：
slaveof 192.168.10.205 6379
 
接着创建redis的数据目录
[root@redis-master redis-3.2.0]# mkdir -p /var/redis/redis
 
然后启动两个节点的redis服务
[root@redis-master redis-3.2.0]# /etc/init.d/redis start
Starting Redis server...
Redis is running...
[root@redis-master redis-3.2.0]# lsof -i:6379
COMMAND     PID USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
redis-ser 17265 root    4u  IPv4  59068      0t0  TCP *:6379 (LISTEN)

2）Keepalived安装（两个节点机都要操作）

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
[root@redis-master ~]# yum -y install openssl openssl-devel
[root@redis-master ~]# cd /usr/local/src/
[root@redis-master src]# wget http://www.keepalived.org/software/keepalived-1.4.0.tar.gz
[root@redis-master src]# tar -zvxf keepalived-1.4.0.tar.gz
[root@redis-master src]# cd keepalived-1.4.0
[root@redis-master keepalived-1.4.0]# ./configure && make && make install
 
文件配置
[root@redis-master keepalived-1.4.0]# mkdir /etc/keepalived
[root@redis-master keepalived-1.4.0]# mkdir /usr/local/keepalived/scripts/ -p
[root@redis-master keepalived-1.4.0]# cp /usr/local/src/keepalived-1.4.0/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/
[root@redis-master keepalived-1.4.0]# cp /usr/local/src/keepalived-1.4.0/keepalived/etc/init.d/keepalived /etc/init.d/
[root@redis-master keepalived-1.4.0]# cp /usr/local/sbin/keepalived /usr/sbin
[root@redis-master keepalived-1.4.0]# cp /usr/local/src/keepalived-1.4.0/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
 
设置开机启动
[root@redis-master keepalived-1.4.0]# chmod +x /etc/init.d/keepalived
[root@redis-master keepalived-1.4.0]# chkconfig --add keepalived
[root@redis-master keepalived-1.4.0]# chkconfig keepalived on

redis主从配置简单说明

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
redis的主从复制实现简单却功能强大，其具有以下特点：
1）一个master支持多个slave连接，slave可以接受其他slave的连接
2）主从同步时，master和slave都是非阻塞的
 
redis主从复制可以用来：
1）data redundancy（数据冗余）
2）slave作为master的扩展，提供一些read-only的服务
3）可以将数据持久化放在slave做，从而提升master性能
 
通过简单的配置slave（master端无需配置），用户就能使用redis的主从复制，即只需在slave端的redis.conf文件中配置下面一行：
slaveof <masterip> <masterport>
表示该redis服务作为slave，masterip和masterport分别为master 的ip和port
 
其他配置：
masterauth <master-password>
如果master设置了安全密码，则此处设置为相应的密码
 
slave-serve-stale-data yes  当slave丢失master或者同步正在进行时，如果发生对slave的服务请求：
slave-serve-stale-data设置为yes则slave依然正常提供服务
slave-serve-stale-data设置为no则slave返回client错误："SYNC with master in progress"
 
repl-ping-slave-period 10
slave发送PINGS到master的时间间隔
 
repl-timeout 60
IO超时时间

3）redis+keepalived配置

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
a）先进行redis-master主节点的高可用配置
[root@redis-master ~]# cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
[root@redis-master ~]# vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived
 
global_defs {
  router_id redis-master
}
 
vrrp_script chk_redis {
  script "/usr/local/keepalived/scripts/redis_check.sh 127.0.0.1 6379"        #监控脚本 
  interval 2                                    #监控时间
  timeout 2                                     #超时时间
  fall 3
}
 
vrrp_instance redis {
  state BACKUP                             
  interface eth0
  lvs_sync_daemon_interface eth0
  virtual_router_id 202
  priority 150                              #权重值                 
  nopreempt                                 #nopreempt：设置不抢占，这里只能设置在state为backup的节点上，而且这个节点的优先级必须比另外节点的高
  advert_int 1
 
  authentication {                          #all node must same             
    auth_type PASS                            #加密   
    auth_pass 1111                            #密码
  }
 
  virtual_ipaddress {
    192.168.10.230                            #VIP地址                 
  }
 
  track_script {
    chk_redis
  }
 
notify_master "/usr/local/keepalived/scripts/redis_master.sh 127.0.0.1 192.168.10.206 6379"
notify_backup "/usr/local/keepalived/scripts/redis_backup.sh 127.0.0.1 192.168.10.206 6379"
notify_fault /usr/local/keepalived/scripts/redis_fault.sh
notify_stop /usr/local/keepalived/scripts/redis_stop.sh
}
 
b）接着进行redis-slave从节点的高可用配置
[root@redis-slave ~]# cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak
[root@redis-slave ~]# vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived
 
global_defs {
  router_id redis-slave
}
 
vrrp_script chk_redis{
  script "/usr/local/keepalived/scripts/redis_check.sh 127.0.0.1 6379"                           
  interval 2                                    
  timeout 2                                    
  fall 3                                   
}
 
vrrp_instance redis {
  state BACKUP                                    
  interface eth0                               
  lvs_sync_daemon_interface eth0                     
  virtual_router_id 202
  priority  100                              
  nopreempt
  advert_int 1
 
  authentication {  
    auth_type PASS                              
    auth_pass 1111                        
  }
 
  virtual_ipaddress {
    192.168.10.230                                    
  }
 
  track_script {
  chk_redis
  }
 
notify_master "/usr/local/keepalived/scripts/redis_master.sh 127.0.0.1 192.168.10.205 6379"
notify_backup "/usr/local/keepalived/scripts/redis_backup.sh 127.0.0.1 192.168.10.205 6379"
notify_fault /usr/local/keepalived/scripts/redis_fault.sh
notify_stop /usr/local/keepalived/scripts/redis_stop.sh
}
 
c）在redis-master和redis-slave两个节点机器上都要创建监控脚本（下面几个脚本，在两个节点上都要同样配置）
首先配置监控脚本
[root@redis-master ~]# vim /usr/local/keepalived/scripts/redis_check.sh
#!/bin/bash
ALIVE=`/usr/local/redis/bin/redis-cli -h $1 -p $2 PING`
LOGFILE="/var/log/keepalived-redis-check.log"
echo "[CHECK]" >> $LOGFILE
date >> $LOGFILE
if [ $ALIVE == "PONG" ]; then :
   echo "Success: redis-cli -h $1 -p $2 PING $ALIVE" >> $LOGFILE 2>&1
    exit 0
else
    echo "Failed:redis-cli -h $1 -p $2 PING $ALIVE " >> $LOGFILE 2>&1
    exit 1
fi
 
需要注意的是：
以下负责运作的关键脚本：
notify_master /usr/local/keepalived/scripts/redis_master.sh
notify_backup /usr/local/keepalived/scripts/redis_backup.sh
notify_fault /usr/local/keepalived/scripts/redis_fault.sh
notify_stop /usr/local/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

   温馨提示：
   以上的keepalived.conf文件中的切换模式设置为nopreempt，意思是：
   不抢占VIP资源，此种模式要是所有的节点都必须设置为state BACKUP模式！
   需要注意无论主备服务器都需要设置为BACKUP，与以往KeepAlived的配置不同，其目的就是防止主服务器恢复后重新抢回VIP，导致Redis切换从而影响稳定。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
接着在redis-master主节点上创建notity_master与notify_backup脚本：
[root@redis-master ~]# vim /usr/local/keepalived/scripts/redis_master.sh
#!/bin/bash
REDISCLI="/usr/local/redis/bin/redis-cli -h $1 -p $3"
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 $2 $3 >> $LOGFILE  2>&1
 
#echo "SLAVEOF $2 cmd can't excute ... " >> $LOGFILE
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
 
[root@redis-master ~]# vim /usr/local/keepalived/scripts/redis_backup.sh
#!/bin/bash
REDISCLI="/usr/local/redis/bin/redis-cli"
LOGFILE="/var/log/keepalived-redis-state.log"
echo "[BACKUP]" >> $LOGFILE
date >> $LOGFILE
echo "Being slave...." >> $LOGFILE 2>&1
echo "Run SLAVEOF cmd ..." >> $LOGFILE 2>&1
$REDISCLI SLAVEOF $2 $3 >> $LOGFILE
sleep 100                                             #延迟100秒以后待数据同步完成后再取消同步状态
exit(0)
 
[root@redis-master ~]# vim /usr/local/keepalived/scripts/redis_fault.sh
#!/bin/bash
LOGFILE="/var/log/keepalived-redis-state.log"
echo "[fault]" >> $LOGFILE
date >> $LOGFILE
 
[root@redis-master ~]# vim /usr/local/keepalived/scripts/redis_stop.sh
#!/bin/bash
LOGFILE="/var/log/keepalived-redis-state.log"
echo "[stop]" >> $LOGFILE
date >> $LOGFILE
 
[root@redis-master ~]# chmod 755 /usr/local/keepalived/scripts/*.sh
[root@redis-master ~]# ll /usr/local/keepalived/scripts/
total 20
-rwxr-xr-x. 1 root root 283 May  7 07:20 redis_backup.sh
-rwxr-xr-x. 1 root root 360 May  7 07:12 redis_check.sh
-rwxr-xr-x. 1 root root 102 May  7 07:22 redis_fault.sh
-rwxr-xr-x. 1 root root 445 May  7 07:16 redis_master.sh
-rwxr-xr-x. 1 root root 101 May  7 07:23 redis_stop.sh
 
将redis-master主节点上的上面5个脚本直接复制到redis-slave节点上即可。
[root@redis-master ~]# rsync -e "ssh -p22" -avpgolr /usr/local/keepalived/scripts/*.sh root@192.168.10.206:/usr/local/keepalived/scripts/
 
到redis-slave从节点上查看脚本：
[root@redis-slave ~]# ll /usr/local/keepalived/scripts/
total 20
-rwxr-xr-x. 1 root root 283 May  7 07:20 redis_backup.sh
-rwxr-xr-x. 1 root root 360 May  7 07:12 redis_check.sh
-rwxr-xr-x. 1 root root 102 May  7 07:22 redis_fault.sh
-rwxr-xr-x. 1 root root 445 May  7 07:16 redis_master.sh
-rwxr-xr-x. 1 root root 101 May  7 07:23 redis_stop.sh
 
d）设置环境变量（两个节点上都要设置）
[root@redis-master ~]# vim /etc/profile
......
export PATH=$PATH:/usr/local/redis/bin
[root@redis-master ~]# source /etc/profile
 
e）启动两个节点上的keepalived服务
[root@redis-master ~]# /etc/init.d/keepalived start
Starting keepalived:                                       [  OK  ]
[root@redis-master ~]# ps -ef|grep keepalived
root     32509     1  0 07:29 ?        00:00:00 keepalived -D
root     32510 32509  0 07:29 ?        00:00:00 keepalived -D
root     32512 32509  0 07:29 ?        00:00:00 keepalived -D
root     32515 32512  0 07:29 ?        00:00:00 keepalived -D
root     32517 32515  0 07:29 ?        00:00:00 /bin/bash /usr/local/keepalived/scripts/redis_backup.sh 127.0.0.1 192.168.10.206 6379
root     32529 14122  0 07:29 pts/1    00:00:00 grep keepalived
 
[root@redis-slave ~]# /etc/init.d/keepalived start
Starting keepalived:                                       [  OK  ]
[root@redis-slave ~]# ps -ef|grep keepalived
root     22277     1  0 07:29 ?        00:00:00 keepalived -D
root     22278 22277  0 07:29 ?        00:00:00 keepalived -D
root     22279 22277  0 07:29 ?        00:00:00 keepalived -D
root     22283 22279  0 07:29 ?        00:00:00 keepalived -D
root     22284 22283  0 07:29 ?        00:00:00 /bin/bash /usr/local/keepalived/scripts/redis_backup.sh 127.0.0.1 192.168.10.205 6379
root     22289 10868  0 07:29 pts/1    00:00:00 grep keepalived
 
查看下redis-master主节点，发现vip资源已经有了
[root@redis-master ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 52:54:00:b1:9c:93 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.205/24 brd 192.168.10.255 scope global eth0
    inet 192.168.10.230/32 scope global eth0
    inet6 fe80::5054:ff:feb1:9c93/64 scope link
       valid_lft forever preferred_lft forever

4）redis+keepalived主从高可用故障切换测试

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
a）分别启动redis-master和redis-slave两个节点的redis和keepalived服务（如上已启动）
   
b）尝试通过VIP连接Redis:
[root@redis-master ~]# redis-cli -h 192.168.10.230 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.205 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.206 INFO|grep role
role:slave
连接成功，Slave也连接上来了。
   
c）尝试插入一些数据：
[root@redis-master ~]# redis-cli -h 192.168.10.230 SET Hello Redis
OK
   
从VIP读取数据
[root@redis-master ~]# redis-cli -h 192.168.10.230 GET Hello
"Redis"
   
从redis-master主节点读取数据
[root@redis-master ~]# redis-cli -h 192.168.10.205 GET Hello
"Redis"
   
从redis-slave从节点读取数据
[root@redis-master ~]# redis-cli -h 192.168.10.206 GET Hello
"Redis"
   
e）然后开始模拟故障产生：
将redis-master主节点上的redis进程杀死：
[root@redis-master ~]# ps -ef|grep redis
root      4500 14122  0 08:04 pts/1    00:00:00 grep redis
root     17265     1  0 04:00 ?        00:00:07 /usr/local/redis/bin/redis-server 0.0.0.0:6379
[root@redis-master ~]# kill -9 17265
[root@redis-master ~]# ps -ef|grep redis
root      4514 14122  0 08:04 pts/1    00:00:00 grep redis
   
查看redis-master主节点上的Keepalived日志
[root@redis-master ~]# tail -f /var/log/keepalived-redis-state.log
OK
[master]
Mon May  7 07:29:17 CST 2018
Being master....
Run SLAVEOF cmd ...
OK Already connected to specified master
Run SLAVEOF NO ONE cmd ...
OK
[fault]
Mon May  7 08:05:00 CST 2018
   
同时redis-slave从节点上的日志显示：
[root@redis-slave ~]# tail -f /var/log/keepalived-redis-state.log
Being slave....
Run SLAVEOF cmd ...
OK
[master]
Mon May  7 08:05:02 CST 2018
Being master....
Run SLAVEOF cmd ...
OK Already connected to specified master
Run SLAVEOF NO ONE cmd ...
OK
   
然后我们可以发现，redis-slave从节点已经接管服务，并且担任Master的角色了。
redis-slave从节点上已经接管过来VIP资源了（大概需要等待2秒左右的时间，vip资源就切过来了）
[root@redis-slave ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 52:54:00:dd:84:6b brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.206/24 brd 192.168.10.255 scope global eth0
    inet 192.168.10.230/32 scope global eth0
    inet6 fe80::5054:ff:fedd:846b/64 scope link
       valid_lft forever preferred_lft forever
   
[root@redis-slave ~]# redis-cli -h 192.168.10.230 INFO|grep role
role:master
[root@redis-slave ~]# redis-cli -h 192.168.10.205 INFO|grep role
Could not connect to Redis at 192.168.10.205:6379: Connection refused
[root@redis-slave ~]# redis-cli -h 192.168.10.206 INFO|grep role
role:master
   
=======================================================================
然后再恢复redis-master主节点的redis进程
[root@redis-master ~]# /etc/init.d/redis start
/var/run/redis.pid exists, process is already running or crashed
Redis is running...
[root@redis-master ~]# rm -f /var/run/redis.pid
[root@redis-master ~]# /etc/init.d/redis start
Starting Redis server...
Redis is running...
[root@redis-master ~]# ps -ef|grep redis
root      4969     1  0 08:08 ?        00:00:00 /usr/local/redis/bin/redis-server 0.0.0.0:6379
root      4977  4976  0 08:08 ?        00:00:00 /bin/bash /usr/local/keepalived/scripts/redis_backup.sh 127.0.0.1 192.168.10.206 6379
root      4987 14122  0 08:08 pts/1    00:00:00 grep redis
[root@redis-master ~]# lsof -i:6379
COMMAND    PID USER   FD   TYPE DEVICE SIZE/OFF NODE NAME
redis-ser 4969 root    4u  IPv4  93698      0t0  TCP *:6379 (LISTEN)
redis-ser 4969 root    6u  IPv4  93709      0t0  TCP 192.168.10.205:43299->192.168.10.206:6379 (ESTABLISHED)
   
查看redis-master上的Keepalived日志
[root@redis-master ~]# tail -f /var/log/keepalived-redis-state.log
OK Already connected to specified master
Run SLAVEOF NO ONE cmd ...
OK
[fault]
Mon May  7 08:05:00 CST 2018
[BACKUP]
Mon May  7 08:08:34 CST 2018
Being slave....
Run SLAVEOF cmd ...
OK
   
查看redis-slave上的Keepalived日志
[root@redis-slave ~]# tail -f /var/log/keepalived-redis-state.log
Being slave....
Run SLAVEOF cmd ...
OK
[master]
Mon May  7 08:05:02 CST 2018
Being master....
Run SLAVEOF cmd ...
OK Already connected to specified master
Run SLAVEOF NO ONE cmd ...
OK
   
[root@redis-master ~]# redis-cli -h 192.168.10.230 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.205 INFO|grep role
role:slave
[root@redis-master ~]# redis-cli -h 192.168.10.206 INFO|grep role
role:master
   
发现redis-master的redis服务再次启动后，redis-master主节点成为salve角色了，redis-slave从节点还是master角色。
当redis-slave节点宕机或redis服务关闭后，redis-master节点再次接管服务成为master角色，如此循环~~
   
关闭redis-slave从节点的reids服务
[root@redis-slave ~]# ps -ef|grep redis
root     15407     1  0 04:00 ?        00:00:10 /usr/local/redis/bin/redis-server 0.0.0.0:6379
root     22900 10868  0 08:11 pts/1    00:00:00 grep redis
[root@redis-slave ~]# kill -9 15407
[root@redis-slave ~]# ps -ef|grep redis
root     22902 10868  0 08:11 pts/1    00:00:00 grep redis
   
查看redis-slave上的Keepalived日志
[root@redis-slave ~]# tail -f /var/log/keepalived-redis-state.log
.......
[stop]          //测试时发现，当redis-slave的redis服务关闭后，还需要重启或关闭keepalived，才能将vip资源漂到redis-master节点上，所以日志里也就会出现这个stop信息
Mon May  7 09:25:03 CST 2018
[BACKUP]
Mon May  7 09:25:04 CST 2018
Being slave....
Run SLAVEOF cmd ...
OK
  
查看redis-master上的Keepalived日志
[root@redis-master ~]# tail -f /var/log/keepalived-redis-state.log
.......
[master]
Mon May  7 09:25:03 CST 2018
Being master....
Run SLAVEOF cmd ...
OK Already connected to specified master
Run SLAVEOF NO ONE cmd ...
OK
  
查看redis-master，发现VIP资源已经接管过来了（如果没有按时切过来的话，只需重启或关闭redis-slave节点那边的keepalived服务即可）
[root@redis-master ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 52:54:00:b1:9c:93 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.205/24 brd 192.168.10.255 scope global eth0
    inet 192.168.10.230/32 scope global eth0
    inet6 fe80::5054:ff:feb1:9c93/64 scope link
       valid_lft forever preferred_lft forever
   
[root@redis-master ~]# redis-cli -h 192.168.10.230 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.205 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.206 INFO|grep role
Could not connect to Redis at 192.168.10.206:6379: Connection refused
   
发现redis-maste节点已经转变为master角色了。
   
同样，当reids-slave节点的redis服务重新启动后，它将成为slave角色。
[root@redis-slave ~]# /etc/init.d/redis start
/var/run/redis.pid exists, process is already running or crashed
Redis is running...
[root@redis-slave ~]# rm -f /var/run/redis.pid
[root@redis-slave ~]# /etc/init.d/redis start
Starting Redis server...
Redis is running...
[root@redis-slave ~]# lsof -i:6379
COMMAND     PID USER   FD   TYPE  DEVICE SIZE/OFF NODE NAME
redis-ser 23244 root    4u  IPv4 3049509      0t0  TCP *:6379 (LISTEN)
redis-ser 23244 root    6u  IPv4 3049513      0t0  TCP dns.kevin.cn:44931->192.168.10.205:6379 (ESTABLISHED)
   
[root@redis-master ~]# redis-cli -h 192.168.10.230 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.205 INFO|grep role
role:master
[root@redis-master ~]# redis-cli -h 192.168.10.206 INFO|grep role
role:slave

ref:https://www.cnblogs.com/kevingrace/p/9001975.html

posted on 2021-08-26 15:32  DoubleLi  阅读(427)  评论(0)  编辑  收藏  举报