CentOS7.6搭建redis4.0.1 cluster集群

1. 操作系统信息：

$ uname -a
　　Linux iZbp11d57wmumnwuihb2czZ 3.10.-957.5..el7.x86_64 # SMP Fri Feb  :: UTC  x86_64 x86_64 x86_64 GNU/Linux
$ cat /proc/version
　　Linux version 3.10.-957.5..el7.x86_64 (mockbuild@kbuilder.bsys.centos.org) (gcc version 4.8.  (Red Hat 4.8.-) (GCC) ) # SMP Fri Feb  :: UTC
$ cat /etc/redhat-release
　　CentOS Linux release 7.6. (Core)

三台ECS机器：172.16.214.160, 172.16.214.161, 172.16.214.163

2. 安裝 GCC 编译工具

$ yum install -y gcc g++ gcc-c++ make

升级所有的包，防止出现版本过久不兼容问题

$ yum -y update

关闭防火墙节点之前需要开放指定端口

systemctl stop firewalld.service

3.1. 安装redis，下载，解压，编译安装

cd /opt
$ wget http://download.redis.io/releases/redis-4.0.1.tar.gz
$ tar xzf redis-4.0..tar.gz
$ cd redis-4.0.
$ make

如果因为上次编译失败，有残留的文件

$ make distclean

172.16.214.160机器上创建redis目录

$ mkdir /opt/redis-4.0./redis-cluster
$ cd /opt/redis-4.0./redis-cluster
$ mkdir

172.16.214.161机器上创建redis目录

$ mkdir /opt/redis-4.0./redis-cluster
$ cd /opt/redis-4.0./redis-cluster
$ mkdir

172.16.214.163机器上创建redis目录

$ mkdir /opt/redis-4.0./redis-cluster
$ cd /opt/redis-4.0./redis-cluster
$ mkdir

每个目录下创建一个redis.conf文件，内容如下（不同的端口和ip需要设置做修改）

#端口号
port 
 
#默认ip为127.0.0.，需要改为其他节点机器可访问的ip，否则创建集群时无法访问对应的端口，无法创建集群
bind 172.16.214.160
 
#redis后台运行
daemonize yes

#访问密码
masterauth passwd123
requirepass passwd123

#pidfile文件
pidfile /var/run/redis_7000.pid
 
#开启集群
cluster-enabled yes
 
#集群的配置，配置文件首次启动自动生成
cluster-config-file nodes_7000.conf
 
#请求超时，默认15秒，可自行设置
cluster-node-timeout     
 
#aof日志开启，有需要就开启，它会每次写操作都记录一条日志
appendonly yes

安装ruby (只需要在其中一台上安装)

$ yum -y install ruby ruby-devel rubygems rpm-build
$ gem install redis

Fetching: redis-4.1.0.gem (100%)
ERROR: Error installing redis:
redis requires Ruby version >= 2.2.2.

这一步有可能会出错，原因是ruby版本太低，解决办法安装高版本ruby

$ yum install curl
$ gpg2 --keyserver hkp://pool.sks-keyservers.net --recv-keys 409B6B1796C275462A1703113804BB82D39DC0E3 7D2BAF1CF37B13E2069D6956105BD0E739499BDB
$ curl -L get.rvm.io | bash -s stable
$ find / -name rvm -print
/usr/local/rvm
/usr/local/rvm/bin/rvm
/usr/local/rvm/scripts/rvm
/usr/local/rvm/lib/rvm
/usr/local/rvm/src/rvm
/usr/local/rvm/src/rvm/bin/rvm
/usr/local/rvm/src/rvm/scripts/rvm
/usr/local/rvm/src/rvm/lib/rvm
$ source /usr/local/rvm/scripts/rvm
$ rvm list known
$ rvm install 2.3.
$ rvm use 2.3.
$ rvm use 2.3. --default
$ ruby --version
$ rvm remove 2.0.
$ gem install redis

设置ruby访问的密码

$  cd /usr/local/rvm/gems/ruby-2.3./gems/redis-4.1./lib/redis/
 
$ vim client.rb

将“:password”后的内容设置为你的密码, 保存退出

3.2 创建用户并设置密码，并切换到该用户身份, 修改redis目录所有权

$ useradd admin 
 
$ passwd admin

$ chown -R admin:admin /opt/redis-4.0.1

$ su admin

4. 为了安全，用admin用户身份启动集群

#第一台机器上执行 3个节点
$ for((i=0;i<=2;i++)); do /opt/redis-4.0.1/src/redis-server /opt/redis-4.0.1/redis-cluster/700$i/redis.conf; done
 
#第二台机器上执行 3个节点
$ for((i=3;i<=5;i++)); do /opt/redis-4.0.1/src/redis-server /opt/redis-4.0.1/redis-cluster/700$i/redis.conf; done
 
#第三台机器上执行 3个节点
$ for((i=6;i<=8;i++)); do /opt/redis-4.0.1/src/redis-server /opt/redis-4.0.1/redis-cluster/700$i/redis.conf; done

检查各 Redis 各个节点启动情况

$ ps -ef | grep redis           //redis是否启动成功
$ netstat -tnlp | grep redis    //监听redis端口

创建集群

$ /opt/redis-4.0./src/redis-trib.rb create --replicas  172.16.214.160: 172.16.214.160: 172.16.214.160: 172.16.214.161: 172.16.214.161: 172.16.214.161: 172.16.214.163: 172.16.214.163: 172.16.214.163:7008

>>> Creating cluster
>>> Performing hash slots allocation on 9 nodes...
Using 4 masters:
172.16.214.160:7000
172.16.214.161:7003
172.16.214.163:7006
172.16.214.160:7001
Adding replica 172.16.214.161:7004 to 172.16.214.160:7000
Adding replica 172.16.214.163:7007 to 172.16.214.161:7003
Adding replica 172.16.214.160:7002 to 172.16.214.163:7006
Adding replica 172.16.214.161:7005 to 172.16.214.160:7001
Adding replica 172.16.214.163:7008 to 172.16.214.160:7000
M: 7c622ac191edd40dd61d9b79b27f6f69d02a5bbf 172.16.214.160:7000
slots:0-4095 (4096 slots) master
M: 44c81c15b01d992cb9ede4ad35477ec853d70723 172.16.214.160:7001
slots:12288-16383 (4096 slots) master
S: 38f03c27af39723e1828eb62d1775c4b6e2c3638 172.16.214.160:7002
replicates f1abb62a8c9b448ea14db421bdfe3f1d8075189c
M: 987965baf505a9aa43e50e46c76189c51a8f17ec 172.16.214.161:7003
slots:4096-8191 (4096 slots) master
S: 6555292fed9c5d52fcf5b983c441aff6f96923d5 172.16.214.161:7004
replicates 7c622ac191edd40dd61d9b79b27f6f69d02a5bbf
S: 2b5ba254a0405d4efde4c459867b15176f79244a 172.16.214.161:7005
replicates 44c81c15b01d992cb9ede4ad35477ec853d70723
M: f1abb62a8c9b448ea14db421bdfe3f1d8075189c 172.16.214.163:7006
slots:8192-12287 (4096 slots) master
S: eb4067373d36d8a8df07951f92794e67a6aac022 172.16.214.163:7007
replicates 987965baf505a9aa43e50e46c76189c51a8f17ec
S: 2919e041dd3d1daf176d6800dcd262f4e727f366 172.16.214.163:7008
replicates 7c622ac191edd40dd61d9b79b27f6f69d02a5bbf
Can I set the above configuration? (type 'yes' to accept): yes

输入yes，然后回车

如何关闭集群：

$ pkill redis

或者逐个关闭

$ for((i=;i<=;i++)); do /opt/redis-4.0./src/redis-cli -c -h 172.16.214.160 -p $i -a passwd123 shutdown; done
 
$ for((i=;i<=;i++)); do /opt/redis-4.0./src/redis-cli -c -h 172.16.214.161 -p $i -a passwd123 shutdown; done
 
$ for((i=;i<=;i++)); do /opt/redis-4.0./src/redis-cli -c -h 172.16.214.163 -p $i -a passwd123 shutdown; done

5.集群验证

参数 -C 可连接到集群，因为 redis.conf 将 bind 改为了ip地址，所以 -h 参数不可以省略，-p 参数为端口号

$ /opt/redis-4.0./src/redis-cli -h 172.16.214.160 -c -p 7000 -a passwd123
172.16.214.160:> set name justfortest
-> Redirected to slot [] located at 172.16.214.161:
OK
172.16.214.161:> get name
"justfortest"
172.16.214.161:>

发现redis get name 重定向，说明集群已经是可用。

检查集群状态

$ /opt/redis-4.0./src/redis-trib.rb check 172.16.214.160:
>>> Performing Cluster Check (using node 172.16.214.160:)
M: 8a78f3b5c1694caecfe9404f131234328481136f 172.16.214.160:
   slots:- ( slots) master
    additional replica(s)
M: d4f15c233b82ec71997f72bb9c04a4fb1f44eea8 172.16.214.161:
   slots:- ( slots) master
    additional replica(s)
M: 729f02e82887746b85d709e36f9b07148e6af442 172.16.214.163:
   slots:- ( slots) master
    additional replica(s)
S: 33a6c6f63ad565e6a9bc76712956d2cbc9819b93 172.16.214.160:
   slots: ( slots) slave
   replicates 729f02e82887746b85d709e36f9b07148e6af442
S: 0860ba2d9cadd23757b2052fe653857fd64e4363 172.16.214.163:
   slots: ( slots) slave
   replicates 8a78f3b5c1694caecfe9404f131234328481136f
S: e5c21b38ed2b6b4870df5f8438259dde0687b363 172.16.214.161:
   slots: ( slots) slave
   replicates 1ec686b61ce3b8a1161382503c62748fc1985553
M: 1ec686b61ce3b8a1161382503c62748fc1985553 172.16.214.160:
   slots:- ( slots) master
    additional replica(s)
S: 37cf628f031d47fc80dd02247d555eb6ab179686 172.16.214.161:
   slots: ( slots) slave
   replicates 8a78f3b5c1694caecfe9404f131234328481136f
S: b586d779805f2f75784bb526ae205cd672534b7a 172.16.214.163:
   slots: ( slots) slave
   replicates d4f15c233b82ec71997f72bb9c04a4fb1f44eea8
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All  slots covered.

列出集群节点

$ /opt/redis-4.0./src/redis-cli -h 172.16.214.160 -c -p 7000 -a passwd123
172.16.214.160:> cluster nodes
d4f15c233b82ec71997f72bb9c04a4fb1f44eea8 172.16.214.161:@ master -    connected -
729f02e82887746b85d709e36f9b07148e6af442 172.16.214.163:@ master -    connected -
33a6c6f63ad565e6a9bc76712956d2cbc9819b93 172.16.214.160:@ slave 729f02e82887746b85d709e36f9b07148e6af442    connected
0860ba2d9cadd23757b2052fe653857fd64e4363 172.16.214.163:@ slave 8a78f3b5c1694caecfe9404f131234328481136f    connected
e5c21b38ed2b6b4870df5f8438259dde0687b363 172.16.214.161:@ slave 1ec686b61ce3b8a1161382503c62748fc1985553    connected
1ec686b61ce3b8a1161382503c62748fc1985553 172.16.214.160:@ master -    connected -
37cf628f031d47fc80dd02247d555eb6ab179686 172.16.214.161:@ slave 8a78f3b5c1694caecfe9404f131234328481136f    connected
b586d779805f2f75784bb526ae205cd672534b7a 172.16.214.163:@ slave d4f15c233b82ec71997f72bb9c04a4fb1f44eea8    connected
8a78f3b5c1694caecfe9404f131234328481136f 172.16.214.160:@ myself,master -    connected -

打印集群信息

$ 172.16.214.160:> cluster info
cluster_state:ok
cluster_slots_assigned:
cluster_slots_ok:
cluster_slots_pfail:
cluster_slots_fail:
cluster_known_nodes:
cluster_size:
cluster_current_epoch:
cluster_my_epoch:
cluster_stats_messages_ping_sent:
cluster_stats_messages_pong_sent:
cluster_stats_messages_sent:
cluster_stats_messages_ping_received:
cluster_stats_messages_pong_received:
cluster_stats_messages_meet_received:
cluster_stats_messages_publish_received:
cluster_stats_messages_received:

6. 为了安全，最后可以设置admin用户身份不能远程登录

chsh admin -s /sbin/nologin