腾讯云，阿里云redis5.0.2集群搭建三主三从，centos7.6

一，本次实验在腾讯云

前提关闭防火墙 想要后期再开 增加防火墙策略

主机3台分别：t1db，t2db，t3db

系统：centos7.6

腾讯云的安全组必须打开就算几台主机在一个安全组也要打开，腾讯比较安全组权限狠低，不开我就等了半天

redis端口：3121，3122

redis集群端口：13121，13122  备注：集群端口为redis端口加10000，不需要配置，但防火墙或者安全组tcp必须打开

redis版本：5.0.2

 

2，安装redis

wget http://download.redis.io/releases/redis-5.0.2.tar.gz

tar -zxvf redis-5.0.2.tar.gz

yum install -y make gcc

make

make install

cd utils/

./install_server.sh

提示默认端口port：6379 我修改为3121

redis文件位置，下面的6379都变成3121，后面我就不说位置在哪了

Port           : 6379
Config file    : /etc/redis/6379.conf
Log file       : /var/log/redis_6379.log
Data dir       : /var/lib/redis/6379
Executable     : /usr/local/bin/redis-server
Cli Executable : /usr/local/bin/redis-cli

增加3122实例启动，这一步应该放在修改3212.conf之后，但既然说安装了 我就放这了

mkdir -p /var/lib/redis/3122

cp /etc/init.d/redis_3122 /etc/init.d/redis_3122

vim /etc/init.d/redis_3122
把里面3121的都改成3122

chkconfig --add redis_3122

启动关闭重启命令

systemctl restart redis_3121        /重启
systemctl restart redis_3122     /重启
systemctl stop redis_3121        /关闭
systemctl stop redis_3122        /关闭
systemctl start redis_3121        /启动
systemctl start redis_3122        /启动
设置了密码，关闭就会无效
可以试用pkill redis-server

启动不了用这种方法
/usr/local/bin/redis-server /etc/redis/3121.conf
/usr/local/bin/redis-server /etc/redis/3122.conf

redis-cli -h ip（例如：127.0.0.1） -p 3121 shutdown  /关闭

查看端口netstat -apn | grep 3121

 

三，3121.conf和3122.conf配置

vi  /etc/redis/3121.conf

允许远程访问

bind 0.0.0.0

protected-mode no

# 绑定服务器IP地址，这个其实0,0,0,0也行 为了安全用本地ip只能本机登录
bind 0.0.0.0

# 绑定端口号，必须修改，以此来区分Redis实例
port 3121

# 后台运行
daemonize yes

# 修改pid进程文件名，以端口号命名
pidfile /var/run/redis_3121.pid

# 修改日志文件名称，以端口号为目录来区分
logfile /var/log/redis_3121.log

# 修改数据文件存放地址，以端口号为目录名来区分
dir /var/lib/redis/3121

# 启用集群
cluster-enabled yes

# 配置每个节点的配置文件，同样以端口号为名称
cluster-config-file nodes-3121.conf

# 配置集群节点的超时时间，可改可不改
cluster-node-timeout 15000

# 启动AOF增量持久化策略
appendonly yes

# 发生改变就记录日志
appendfsync always

#设置redis访问密码 ，测试中所以没用，为了安全生产环境请加上
requirepass ssxx 
   
#设置redis集群间的访问密码，同上面密码一致，测试中所以没用，为了安全生产环境请加上
masterauth ssxx    

把3121.conf复制命名3122.conf

vi  /etc/redis/3122.conf

把3121都修改成3122

好了这里第一台主机t1db的就配置完成  ，启动

systemctl start redis_3121        /启动
systemctl start redis_3122

t2db和t3db一样 安装

然后把3121.conf和3122.conf的配置丢到， t2db和t3db主机里只需要修改3121.conf和3122.conf绑定ip

bind 0.0.0.0

 

bind 0.0.0.0

三台主机6个实例都启动

四，创建集群

建立集群时

Waiting for the cluster to join
..........................................................如果这一步等的太久,这么多点点 点 大概率是防火墙或者安全组没开集群端口

 

如果建立集群失败，删除6给实例的文件/var/lib/redis/3122和/var/lib/redis/3121这里的文件目下的

dump.rdb  和nodes-3122.conf

dump.rdb 和 nodes-3121.conf   

无密码建立

redis-cli --cluster create 172.16.0.8:3121 172.16.0.8:3122 172.16.0.9:3121 172.16.0.9:3122 172.16.0.10:3121 172.16.0.10:3122 --cluster-replicas 1

 

有密码建立

redis-cli -a ssxx --cluster create 172.16.0.8:3121 172.16.0.8:3122 172.16.0.9:3121 172.16.0.9:3122 172.16.0.10:3121 172.16.0.10:3122 --cluster-replicas 1

下面是无密码建立后的显示

>>> Performing hash slots allocation on 6 nodes...
Master[0] -> Slots 0 - 5460
Master[1] -> Slots 5461 - 10922
Master[2] -> Slots 10923 - 16383
Adding replica 172.16.0.9:3122 to 172.16.0.8:3121
Adding replica 172.16.0.8:3122 to 172.16.0.9:3121
Adding replica 172.16.0.10:3122 to 172.16.0.10:3121
>>> Trying to optimize slaves allocation for anti-affinity
[OK] Perfect anti-affinity obtained!
M: 1c708a6b01f860835c209e8881d3bc0934734076 172.16.0.8:3121
   slots:[0-5460] (5461 slots) master
S: 2d46d434da5a196fd435f5a4ffb09b9cefbc8f58 172.16.0.8:3122
   replicates 28dc96a888d087c20306231d5713672da0ec8951
M: 28dc96a888d087c20306231d5713672da0ec8951 172.16.0.9:3121
   slots:[5461-10922] (5462 slots) master
S: 8696e44653ae48dc85209face86e02ba0de65e56 172.16.0.9:3122
   replicates 4716bd5554eb9514f5ffba973682ec4bf302f7e9
M: 4716bd5554eb9514f5ffba973682ec4bf302f7e9 172.16.0.10:3121
   slots:[10923-16383] (5461 slots) master
S: 7317a02362de2199e32b3c570f93149041882e6d 172.16.0.10:3122
   replicates 1c708a6b01f860835c209e8881d3bc0934734076
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 172.16.0.8:3121)
M: 1c708a6b01f860835c209e8881d3bc0934734076 172.16.0.8:3121
   slots:[0-5460] (5461 slots) master
   1 additional replica(s)
S: 7317a02362de2199e32b3c570f93149041882e6d 172.16.0.10:3122
   slots: (0 slots) slave
   replicates 1c708a6b01f860835c209e8881d3bc0934734076
M: 4716bd5554eb9514f5ffba973682ec4bf302f7e9 172.16.0.10:3121
   slots:[10923-16383] (5461 slots) master
   1 additional replica(s)
M: 28dc96a888d087c20306231d5713672da0ec8951 172.16.0.9:3121
   slots:[5461-10922] (5462 slots) master
   1 additional replica(s)
S: 8696e44653ae48dc85209face86e02ba0de65e56 172.16.0.9:3122
   slots: (0 slots) slave
   replicates 4716bd5554eb9514f5ffba973682ec4bf302f7e9
S: 2d46d434da5a196fd435f5a4ffb09b9cefbc8f58 172.16.0.8:3122
   slots: (0 slots) slave
   replicates 28dc96a888d087c20306231d5713672da0ec8951
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

到此处成功

修改好配置可以继续建立集群

五，查看集群状态

redis-cli --cluster check 172.16.0.8:3121

172.16.0.8:3121 (1c708a6b...) -> 0 keys | 5461 slots | 1 slaves.
172.16.0.10:3121 (4716bd55...) -> 0 keys | 5461 slots | 1 slaves.
172.16.0.9:3121 (28dc96a8...) -> 0 keys | 5462 slots | 1 slaves.
[OK] 0 keys in 3 masters.
0.00 keys per slot on average.
>>> Performing Cluster Check (using node 172.16.0.8:31621)
M: 1c708a6b01f860835c209e8881d3bc0934734076 172.16.0.8:31621
   slots:[0-5460] (5461 slots) master
   1 additional replica(s)
S: 7317a02362de2199e32b3c570f93149041882e6d 172.16.0.12:31622
   slots: (0 slots) slave
   replicates 1c708a6b01f860835c209e8881d3bc0934734076
M: 4716bd5554eb9514f5ffba973682ec4bf302f7e9 172.16.0.12:31621
   slots:[10923-16383] (5461 slots) master
   1 additional replica(s)
M: 28dc96a888d087c20306231d5713672da0ec8951 172.16.0.9:31621
   slots:[5461-10922] (5462 slots) master
   1 additional replica(s)
S: 8696e44653ae48dc85209face86e02ba0de65e56 172.16.0.9:31622
   slots: (0 slots) slave
   replicates 4716bd5554eb9514f5ffba973682ec4bf302f7e9
S: 2d46d434da5a196fd435f5a4ffb09b9cefbc8f58 172.16.0.8:31622
   slots: (0 slots) slave
   replicates 28dc96a888d087c20306231d5713672da0ec8951
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

 

六，使用集群，随意一个端口实例都行

[root@t1db ~]# redis-cli -c -h 172.16.0.8 -p 3121
172.16.0.8:3121> set a test
-> Redirected to slot [15495] located at 172.16.0.10:3121
OK
172.16.0.10:3121> get a
"test"
172.16.0.10:31621> exit
[root@t1db ~]# redis-cli -c -h 172.16.0.8 -p 3122
172.16.0.8:3122> get a
-> Redirected to slot [15495] located at 172.16.0.10:3121
"test"
172.16.0.12:3121> exit
[root@t1db ~]# redis-cli -c -h 172.16.0.9 -p 3122 
172.16.0.9:3122> get a
-> Redirected to slot [15495] located at 172.16.0.10:3121
"test"
172.16.0.10:3121> exit