redis系列（五）：搭建redis-cluster集群

1、为什么要用redis-cluster

a、并发要求
　　redis官方声称可以达到10万每秒，但是如果业务需要每秒100万条呢？
b、数据量太大
　　一台服务器的内存正常是16-256G，如果业务需要500G内存怎么办？

2、搭建redis-cluster

针对上述问题，redis-cluster集群就提供了很好的解决方案。

（1）、先准备环境，开启多个redis实例

[root@localhost redis_conf]# ls
redis-.conf  redis-.conf  redis-.conf
redis-.conf  redis-.conf  redis-.conf
[root@localhost redis_conf]# 

暂时准备了6个配置文件，

daemonize yes
port
logfile ./data//redis.log #日志存放位置
dir ./data/  # 数据存放位置
dbfilename  dbmp.rdb # 数据文件名称
cluster-enabled yes # 开启集群模式
cluster-config-file nodes-.conf # 集群内部的配置文件
cluster-require-full-coverage no # redis cluster需要16384个slot都正常的时候才能对外提供服务，换句话说，只要任何一个slot异常那么整个cluster不对外提供服务。 因此生产环境一般为no

上面的配置为redis-7000.conf的配置文件内容，其他配置文件内容相同，只是将7000全部改成对应的端口。

可以使用如下命令快速生成：

sed "s/7000/7001/g" redis-.conf > redis-.conf

通过sed命令将redis-7000.conf中的7000修改成7001，然后写入redis-7001.conf文件中

每个节点仅仅是端口的不同。

注意：还要确保配置中的日志以及数据存放文件夹存在。

比如我的：

[root@localhost redis_conf]# mkdir -p data/{,,,,,}
[root@localhost redis_conf]# tree
.
├── data
│   ├──
│   ├──
│   ├──
│   ├──
│   ├──
│   └──
├── redis-.conf
├── redis-.conf
├── redis-.conf
├── redis-.conf
├── redis-.conf
└── redis-.conf

 directories,  files

(2)、运行redis实例

[root@localhost redis_conf]# redis-server redis-.conf
[root@localhost redis_conf]# redis-server redis-.conf
[root@localhost redis_conf]# redis-server redis-.conf
[root@localhost redis_conf]# redis-server redis-.conf
[root@localhost redis_conf]# redis-server redis-.conf
[root@localhost redis_conf]# redis-server redis-.conf

查看是否已经启动

[root@localhost redis_conf]# ps -ef | grep redis
root              : ?        :: redis-server *: [cluster]
root              : ?        :: redis-server *: [cluster]
root              : ?        :: redis-server *: [cluster]
root              : ?        :: redis-server *: [cluster]
root              : ?        :: redis-server *: [cluster]
root              : ?        :: redis-server *: [cluster]
root           : pts/    :: grep --color=auto redis
[root@localhost redis_conf]# 

此时集群还用不了，可以登录redis查看

[root@localhost redis_conf]# redis-cli -p
127.0.0.1:> set name felixi
(error) CLUSTERDOWN Hash slot not served
127.0.0.1:> 

(3)、创建redis-cluster

a、准备ruby环境

下载，编译，安装ruby （ruby官网地址）

、下载（个人用的当前的最新版本2.6.0）
wget https://cache.ruby-lang.org/pub/ruby/2.6/ruby-2.6.0.tar.gz
、解压，安装
tar -zxvf ruby-2.6..tar.gz
cd ruby-2.6.
./configure --prefix=/opt/ruby/
make && make install
、添加环境变量
export PATH=/opt/ruby/bin/:$PATH # 将这句添加到./bashrc和/etc/profile文件末尾。
source ./bashrc /etc/profile # 加载一下

b、查看是否已经安装

[root@localhost ~]# gem -v
3.0.
[root@localhost ~]# 

c、下载安装ruby操作redis的模块包

[root@localhost ~]# gem install redis
Successfully installed redis-4.1.
Parsing documentation for redis-4.1.
Done installing documentation for redis after  seconds
 gem installed
[root@localhost ~]# 

7、启动集群

我的redis版本是5.0.2，使用如下方式启动

redis-cli --cluster create 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: --cluster-replicas 

其他旧版本可能需要如下命令：注意(redis-trib.rb可能找不到，可以通过find / -name redis-trib.rb来查找)

/opt/redis-4.0./src/redis-trib.rb create --replicas  127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1:

8、出现如下说明启动成功

[root@localhost ~]# redis-cli --cluster create 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: 127.0.0.1: --cluster-replicas
>>> Performing hash slots allocation on  nodes...
Master[] -> Slots  -
Master[] -> Slots  -
Master[] -> Slots  -
Adding replica 127.0.0.1: to 127.0.0.1:
Adding replica 127.0.0.1: to 127.0.0.1:
Adding replica 127.0.0.1: to 127.0.0.1:
>>> Trying to optimize slaves allocation for anti-affinity
[WARNING] Some slaves are in the same host as their master
M: 511bc46a1da42e3964ce41f48234076bd5743baf 127.0.0.1:
   slots:[-] ( slots) master
M: bb029209525fb9aa2eeaa8c27182065ca1b29457 127.0.0.1:
   slots:[-] ( slots) master
M: 72b6ad9d36ba9cc7770bfaf6899c8e3262c677b0 127.0.0.1:
   slots:[-] ( slots) master
S: 25c00188cc8b58f5442ac674647389ab4f9206e9 127.0.0.1:
   replicates 511bc46a1da42e3964ce41f48234076bd5743baf
S: bd093c559ebbc54fead2da9200d6f0c10a90bc87 127.0.0.1:
   replicates bb029209525fb9aa2eeaa8c27182065ca1b29457
S: 3c0031255f960fbc15bbf9a78dacd7427ac24115 127.0.0.1:
   replicates 72b6ad9d36ba9cc7770bfaf6899c8e3262c677b0
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 127.0.0.1:)
M: 511bc46a1da42e3964ce41f48234076bd5743baf 127.0.0.1:
   slots:[-] ( slots) master
    additional replica(s)
M: 72b6ad9d36ba9cc7770bfaf6899c8e3262c677b0 127.0.0.1:
   slots:[-] ( slots) master
    additional replica(s)
M: bb029209525fb9aa2eeaa8c27182065ca1b29457 127.0.0.1:
   slots:[-] ( slots) master
    additional replica(s)
S: bd093c559ebbc54fead2da9200d6f0c10a90bc87 127.0.0.1:
   slots: ( slots) slave
   replicates bb029209525fb9aa2eeaa8c27182065ca1b29457
S: 25c00188cc8b58f5442ac674647389ab4f9206e9 127.0.0.1:
   slots: ( slots) slave
   replicates 511bc46a1da42e3964ce41f48234076bd5743baf
S: 3c0031255f960fbc15bbf9a78dacd7427ac24115 127.0.0.1:
   slots: ( slots) slave
   replicates 72b6ad9d36ba9cc7770bfaf6899c8e3262c677b0
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All  slots covered.
[root@localhost ~]# 

9、查看主从状态

redis-cli -p   info replication

结果如下：

[root@localhost ~]# redis-cli -p   info replication
# Replication
role:master
connected_slaves:
slave0:ip=127.0.0.1,port=,state=online,offset=,lag=
master_replid:2b20fa4941542dc58d9fc2c32fe6f3dbb6cce72b
master_replid2:
master_repl_offset:
second_repl_offset:-
repl_backlog_active:
repl_backlog_size:
repl_backlog_first_byte_offset:
repl_backlog_histlen:
[root@localhost ~]# redis-cli -p   info replication
# Replication
role:slave
master_host:127.0.0.1
master_port:
master_link_status:up
master_last_io_seconds_ago:
master_sync_in_progress:
slave_repl_offset:
slave_priority:
slave_read_only:
connected_slaves:
master_replid:ad11aa24aa797d942a54550ee77ab0e185e9d92c
master_replid2:
master_repl_offset:
second_repl_offset:-
repl_backlog_active:
repl_backlog_size:
repl_backlog_first_byte_offset:
repl_backlog_histlen:
[root@localhost ~]# 

9、向redis集群写入数据，查看数据流向

redis-cli -p 7000 -c  #这里会将key自动的重定向，放到某一个节点的slot槽位中

效果如下：

[root@localhost ~]# redis-cli -p  -c
127.0.0.1:> keys *
(empty list or set)
127.0.0.1:>
127.0.0.1:> set name felix
-> Redirected to slot [] located at 127.0.0.1:
OK
127.0.0.1:> keys *
) "name"
127.0.0.1:> 

至此，集群就搭建好了