Linux环境下Redis 集群部署

1.单机Redis部署

Linux下redis安装并设置开机自启

2.Redis 集群配置

具体是参考这篇文章（由于在搭建过程中仍遇到很多问题，所以在其基础上补充）：
Linux下redis集群搭建与部署

2.1 创建redis集群安装目录

cd /home/xh/hadoop/
mkdir myredis
cd myredis
mkdir 7000 7001 7002 7003 7004 7005

2.2 将redis单机部署目录下的redis.confi文件复制到每个目录下

cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7000
cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7001
cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7002
cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7003
cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7004
cp /home/xh/hadoop/redis-3.2.5/redis.conf /home/xh/hadoop/myredis/7005

2.3 修改每个文件夹下的redis.conf

#例如
vim myredis/7000/redis.conf

#修改如下
#端口号
port 7000
#后台启动
daemonize yes
#开启集群
cluster-enabled yes
#集群节点配置文件（注意需要跟端口号一样）
cluster-config-file nodes-7000.conf
#数据文件存放位置(注意此处的路径需要自己创建好)
dir /usr/local/redis/redis-cluster/7000/data/
#集群连接超时时间
cluster-node-timeout 5000
#进程pid的文件位置 （注意需要跟端口号一样）
pidfile /var/run/redis-7000.pid
#开启aof
appendonly yes
#aof文件路径
appendfilename “appendonly-7000.aof”
#rdb文件路径
dbfilename dump-7000.rdb

注意：需要补充如下设置，否则java连接报错

#保护模式设为no
protected-mode no
#绑定ip
bind 192.168.240.128

2.4 修改完六个配置内容后开始启动

cd /usr/local/redis

#这里闲一个个启动麻烦的话可以自己配置一个配置文件启动
bin/redis-server myredis/7000/redis.conf
bin/redis-server myredis/7001/redis.conf
bin/redis-server myredis/7002/redis.conf
bin/redis-server myredis/7003/redis.conf
bin/redis-server myredis/7004/redis.conf
bin/redis-server myredis/7005/redis.conf

2.5 启动完后查看进程

[root@hadoop myredis]# ps -ef|grep redis
#显示有六个则是启动成功
root 63262 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7000 [cluster]
root 63264 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7001 [cluster]
root 63266 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7002 [cluster]
root 63270 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7003 [cluster]
root 63276 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7004 [cluster]
root 63278 1 0 11:08 ? 00:00:00 redis-server 192.168.240.128:7005 [cluster]
root 63287 62960 0 11:08 pts/2 00:00:00 grep redis

2.6 建集群

到这里只是启动了六个单进程的redis，开始创建集群，先安装好ruby

yum install ruby rubygems -y

使用gem要先镜像一下

#这里需要镜像一下
gem sources --add https://gems.ruby-china.com/ --remove https://rubygems.org/

#确保镜像成功
[root@hadoop myredis]# gem sources -l

*** CURRENT SOURCES ***

https://gems.ruby-china.com/

然后执行连接ruby-redis

[root@hadoop myredis]# gem install redis

上面步骤有可能报错查了资料发现是版本太低
解决办法是 先安装rvm，再把ruby版本提升

ERROR: Error installing redis:
redis requires Ruby version >= 2.3.0.

ruby版本提升（若没有报错不需要进行此步骤）

#安装curl
sudo yum install curl

#安装rvm
curl -L get.rvm.io | bash -s stable

(如出现错误参考: http://blog.csdn.net/qq_30242987/article/details/99727838)

source /usr/local/rvm/scripts/rvm

#查看rvm库中已知的ruby版本
rvm list known

#安装一个ruby版本
rvm install 2.4.0

#使用一个ruby版本
rvm use 2.4.0

#卸载一个已知版本
rvm remove 2.3.0

#查看版本
ruby --version

#再安装redis就可以了
gem install redis

创建集群

#进入src下面
cd /home/xh/hadoop/redis-3.2.5/src

# 创建集群
[root@hadoop src]# ./redis-trib.rb create --replicas 1 192.168.240.128:7000 192.168.240.128:7001 192.168.240.128:7002 192.168.240.128:7003 192.168.240.128:7004 192.168.240.128:7005
>>> Creating cluster
>>> Performing hash slots allocation on 6 nodes...
Using 3 masters:
192.168.240.128:7000
192.168.240.128:7001
192.168.240.128:7002
Adding replica 192.168.240.128:7003 to 192.168.240.128:7000
Adding replica 192.168.240.128:7004 to 192.168.240.128:7001
Adding replica 192.168.240.128:7005 to 192.168.240.128:7002
M: b553b5ed57b5d152dc1819c1818e61eba77b867f 192.168.240.128:7000
   slots:0-5460 (5461 slots) master
M: d3bb8e329319f90d42f3a6163d0dcb2059cbeb47 192.168.240.128:7001
   slots:5461-10922 (5462 slots) master
M: 40a44e92c9644e0366abb3aa2f18222b7d255b93 192.168.240.128:7002
   slots:10923-16383 (5461 slots) master
S: 330be48eaa9bd676720738f0efe8007b2421944e 192.168.240.128:7003
   replicates b553b5ed57b5d152dc1819c1818e61eba77b867f
S: ba1195fe77df429079e34cb6c296878b1d0c7dcb 192.168.240.128:7004
   replicates d3bb8e329319f90d42f3a6163d0dcb2059cbeb47
S: 674176635075d82369ebba73060eadf8e4e1d701 192.168.240.128:7005
   replicates 40a44e92c9644e0366abb3aa2f18222b7d255b93
Can I set the above configuration? (type 'yes' to accept): yes
>>> Nodes configuration updated
>>> Assign a different config epoch to each node
>>> Sending CLUSTER MEET messages to join the cluster
Waiting for the cluster to join.....
>>> Performing Cluster Check (using node 192.168.240.128:7000)
M: b553b5ed57b5d152dc1819c1818e61eba77b867f 192.168.240.128:7000
   slots:0-5460 (5461 slots) master
   1 additional replica(s)
M: d3bb8e329319f90d42f3a6163d0dcb2059cbeb47 192.168.240.128:7001
   slots:5461-10922 (5462 slots) master
   1 additional replica(s)
S: ba1195fe77df429079e34cb6c296878b1d0c7dcb 192.168.240.128:7004
   slots: (0 slots) slave
   replicates d3bb8e329319f90d42f3a6163d0dcb2059cbeb47
M: 40a44e92c9644e0366abb3aa2f18222b7d255b93 192.168.240.128:7002
   slots:10923-16383 (5461 slots) master
   1 additional replica(s)
S: 330be48eaa9bd676720738f0efe8007b2421944e 192.168.240.128:7003
   slots: (0 slots) slave
   replicates b553b5ed57b5d152dc1819c1818e61eba77b867f
S: 674176635075d82369ebba73060eadf8e4e1d701 192.168.240.128:7005
   slots: (0 slots) slave
   replicates 40a44e92c9644e0366abb3aa2f18222b7d255b93
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

到这里集群就安装好了，测试一下

[xh@hadoop myredis]$ redis-cli -c -h 192.168.240.128 -p 7000
192.168.240.128:7000> set name frank
-> Redirected to slot [5798] located at 192.168.240.128:7001
OK
192.168.240.128:7001> get name
"frank"

在启动时会发现一个个启动太麻烦，这里配置一下启动关闭文件

#进入安装redis的路径
cd /home/xh/hadoop/myredis
 
#编写关闭文件
vi stop-all.sh
 
#加入下面内容，编辑完之后按esc键输入 :wq 退出保存
#这里得找到自己配置启动得地方，我得是在bin下面
redis-cli -p 7000 -h 192.168.240.128 shutdown
redis-cli -p 7001 -h 192.168.240.128 shutdown
redis-cli -p 7002 -h 192.168.240.128 shutdown
redis-cli -p 7003 -h 192.168.240.128 shutdown
redis-cli -p 7004 -h 192.168.240.128 shutdown
redis-cli -p 7005 -h 192.168.240.128 shutdown
 
#赋值权限
chmod u+x stop-all.sh
 
 
#编写启动文件
vi start-all.sh
 
#加入下面内容，编辑完之后按esc键输入 :wq 退出保存
#这里得找到自己配置启动得地方，我得是在bin下面
redis-server myredis/7000/redis.conf
redis-server myredis/7001/redis.conf
redis-server myredis/7002/redis.conf
redis-server myredis/7003/redis.conf
redis-server myredis/7004/redis.conf
redis-server myredis/7005/redis.conf
 
#赋值权限
chmod u+x start-all.sh

其中在java连接时出现问题，通过参考如下文章解决：
1.本地虚拟机搭建的Redis集群，Jedis可以连接，JedisCluster连接不上
2.Redis错误：[ERR] Sorry, can‘t connect to node 192.168.10.3:6379
3.彻底解决：[ERR] Node is not empty. Either the node already knows other nodes