redis主从配置+哨兵模式

1.搭建redis主从,一个master两个slave,加一个哨兵监听(sentinel),可以新建三个虚拟机,模拟环境,我的电脑没那么多虚拟机,就在一台虚拟机上弄的.

2.安装redis,如果是三台机器,安装redis的步骤一样.

3.安装完redis后,复制同样的文件两份用于当slave

#复制两份redis文件,当从redis(端口6379为主redis,端口6380,6381两个为从redis)[root@localhost src]# cp -r /usr/redis/redis-3.2.4 /usr/redis/redis-6380
[root@localhost src]# cp -r /usr/redis/redis-3.2.4 /usr/redis/redis-6381

4.修改两份 redis.conf 配置文件

[root@localhost src]# vi /usr/redis/redis-6380/redis.conf
[root@localhost src]# vi /usr/redis/redis-6381/redis.conf

redis.conf 配置文件中主要改的地方(以/usr/redis/redis-6380/redis.conf文件为例)

1.bind 192.168.200.108(改为虚拟机的ip)
2.port 6380(端口号)
3.daemonize yes(守护进程)
4.pidfile "/var/run/redis_6380.pid"(进程的目录)
5.slaveof 192.168.200.108 6380(设置为此redis为slave,端口号是6380)
6.slave-read-only yes(只读)

5.修改两份 sentinel.conf 配置文件

[root@localhost src]# vi /usr/redis/redis-6380/sentinel.conf
[root@localhost src]# vi /usr/redis/redis-6381/sentinel.conf

sentinel.conf 配置文件修改的地方

1.port 26379(端口号,可以使用默认端口)
2.sentinel monitor master 192.168.200.108 6379 2(主redis的名称,ip和端口)

6.在主redis配置哨兵监听

[root@localhost src]# vi /usr/redis/redis-3.2.4/sentinel.conf

#修改要监听的主机ip和端口
sentinel monitor mymaster 192.168.200.108 6379 1

7.在/etc/init.d文件中复制两份redis启动脚本(脚本内容见之前写的 启动redis脚本 文章)

修改端口号和文件路径(以redis-6380为例)

8.启动三台redis

启动哨兵

[root@localhost ~]# ps -ef|grep redis
root      3403     1  0 20:28 ?        00:00:00 /usr/redis/redis-3.2.4/src/redis-server 192.168.200.108:6379
root      3415     1  0 20:28 ?        00:00:00 /usr/redis/redis-6380/src/redis-server 192.168.200.108:6380
root      3427     1  0 20:28 ?        00:00:00 /usr/redis/redis-6381/src/redis-server 192.168.200.108:6381
root      3432  3373  0 20:28 pts/1    00:00:00 grep redis
[root@localhost ~]# cd /usr/redis/redis-3.2.4
[root@localhost redis-3.2.4]# src/redis-sentinel sentinel.conf &
[1] 3443
[root@localhost redis-3.2.4]# 3443:X 02 Mar 20:30:29.076 * Increased maximum number of open files to 10032 (it was originally set to 1024).
                _._
           _.-``__ ''-._
      _.-``    `.  `_.  ''-._           Redis 3.2.4 (00000000/0) 64 bit
  .-`` .-```.  ```\/    _.,_ ''-._
 (    '      ,       .-`  | `,    )     Running in sentinel mode
 |`-._`-...-` __...-.``-._|'` _.-'|     Port: 26379
 |    `-._   `._    /     _.-'    |     PID: 3443
  `-._    `-._  `-./  _.-'    _.-'
 |`-._`-._    `-.__.-'    _.-'_.-'|
 |    `-._`-._        _.-'_.-'    |           http://redis.io
  `-._    `-._`-.__.-'_.-'    _.-'
 |`-._`-._    `-.__.-'    _.-'_.-'|
 |    `-._`-._        _.-'_.-'    |
  `-._    `-._`-.__.-'_.-'    _.-'
      `-._    `-.__.-'    _.-'
          `-._        _.-'
              `-.__.-'                                               

3443:X 02 Mar 20:30:29.104 # WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.
3443:X 02 Mar 20:30:29.104 # Sentinel ID is ec55f3b5cca3a78aae0ad0fe94c78d7612221e0b
3443:X 02 Mar 20:30:29.104 # +monitor master master 192.168.200.108 6379 quorum 1
3443:X 02 Mar 20:30:29.104 # +monitor master mymaster 127.0.0.1 6379 quorum 2
3443:X 02 Mar 20:30:59.106 # +sdown master mymaster 127.0.0.1 6379

9.查看三台redis的从属关系

[root@localhost redis-3.2.4]# ps -ef|grep redis
root      3677     1  0 21:18 ?        00:00:02 /usr/redis/redis-3.2.4/src/redis-server 192.168.200.108:6379
root      3777     1  0 21:35 ?        00:00:00 /usr/redis/redis-6380/src/redis-server 192.168.200.108:6380
root      3791     1  0 21:35 ?        00:00:00 /usr/redis/redis-6381/src/redis-server 192.168.200.108:6381
root      3804  3730  0 21:37 pts/3    00:00:00 src/redis-sentinel *:26379 [sentinel]
root      3810  3730  0 21:38 pts/3    00:00:00 grep redis
[root@localhost redis-3.2.4]# cd /usr/redis/redis-3.2.4/src
[root@localhost src]# ./redis-cli -h 192.168.200.108 -p 6379
192.168.200.108:6379> info replication --->查看当前redis信息
# Replication
role:master  --->说明是主redis
connected_slaves:2
slave0:ip=192.168.200.108,port=6380,state=online,offset=6165957,lag=0  --->从redis端口和ip 192.168.200.108 6380
slave1:ip=192.168.200.108,port=6381,state=online,offset=6165667,lag=0  --->从redis端口和ip 192.168.200.108 6381
master_repl_offset:6165957
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:5117382
repl_backlog_histlen:1048576
[root@localhost src]# ./redis-cli -h 192.168.200.108 -p 6380  --->连接192.168.200.108 6380查看信息
192.168.200.108:6380> info replication
# Replication
role:slave  --->从redis
master_host:192.168.200.108  --->主redis ip
master_port:6379  --->主redis端口
master_link_status:up
master_last_io_seconds_ago:0
master_sync_in_progress:0
slave_repl_offset:4430
slave_priority:100
slave_read_only:1
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0
192.168.200.108:6380>
[root@localhost src]# ./redis-cli -h 192.168.200.108 -p 6381  --->连接192.168.200.108 6381查看信息
192.168.200.108:6381> info replication
# Replication
role:slave  --->从redis
master_host:192.168.200.108
master_port:6379
master_link_status:up
master_last_io_seconds_ago:1
master_sync_in_progress:0
slave_repl_offset:73275
slave_priority:100
slave_read_only:1
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

10.当把主redis kill 后,哨兵会监听会master,当发现master挂了后,哨兵开始故障恢复,选取一个从redis升级为主库

常用命令

slaveof no one : 手动设置从redis,从属服务器关闭复制功能，并从从属服务器转变回主服务器，原来同步所得的数据集不会被丢弃

192.168.200.108:6380> slaveof no one
OK

slaveof host port : 将当前服务器转变为指定服务器的从属服务器(slave server)。

192.168.200.108:6381> slaveof 192.168.200.108 6379
OK

参考资源：http://blog.csdn.net/gqtcgq/article/details/50273431