多台宿主机使用docker部署redis集群

最新推荐文章于 2025-06-26 15:27:11 发布

EHOIST

最新推荐文章于 2025-06-26 15:27:11 发布

阅读量3.8k

点赞数 3

CC 4.0 BY-SA版权

分类专栏： docker 文章标签： docker redis 集群多宿主机

本文链接：https://blog.youkuaiyun.com/EHOIST/article/details/88262250

docker 专栏收录该内容

3 篇文章

订阅专栏

本文详细介绍了如何在Centos7.5环境下利用Docker部署一个由六台服务器组成的Redis集群，涵盖了Docker环境的搭建、镜像制作与上传、Ansible自动化部署、集群搭建及验证等关键步骤。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

前言

在Centos7.5上多台宿主机使用docker部署redis集群的过程，redis集群一般需要6台redis服务器,使用docker可以节省服务器资源.第一次发博客，有什么不对的地方还请大家指出。

集群架构

主机名	IP	redis集群ip(并非为redis容器的ip)
test（部署用）	192.168.0.35
redis1	192.168.0.71	192.168.0.71:6379 192.168.0.71:6380
redis2	192.168.0.72	192.168.0.72:6379 192.168.0.72:6380
redis3	192.168.0.73	192.168.0.73:6379 192.168.0.73:6380

主	从
redis1: 6379	redis2 : 6380
redis2 : 6379	redis3 : 6380
redis3 : 6379	redis1 : 6380

docker环境部署

安装docker

[root@test ~]# yum -y install docker-engine

启动docker

[root@test ~]# systemctl start docker

我的docker版本

[root@lianxi-0001 ~]# docker version 
Client:
 Version:      1.12.1
 API version:  1.24
 Go version:   go1.6.3
 Git commit:   23cf638
 Built:        
 OS/Arch:      linux/amd64

下载官方centos镜像

[root@test ~]# docker pull centos

制作redis镜像，wokdir下需要redis-4.0.8源码包解压后的文件夹，可用的repo文件和Dockerfile

[root@test ~]# mkdir workdir
[root@test ~]# ls workdir/
CentOS-Base.repo  Dockerfile  local.repo  redis-4.0.8
[root@test ~]# vim workdir/Dockerfile
FROM centos
RUN rm -rf /etc/yum.repos.d/*
ADD *.repo /etc/yum.repos.d/
ADD redis-4.0.8 /root/redis-4.0.8
WORKDIR /root/redis-4.0.8
RUN yum -y install make  gcc
RUN make && make install
RUN mkdir -p /var/log/redis /var/lib/redis/ /etc/redis/
CMD ["redis-server", "/etc/redis/redis.conf"]
[root@test ~]# docker build -t redis workdir/

制作私有镜像仓库并上传redis镜像

[root@test ~]# vim /etc/docker/daemon.json
{
   "insecure-registries": ["192.168.0.35:5000"]
}
[root@test ~]# systemctl restart docker
[root@test ~]# docker pull registry
[root@test ~]# docker run -itd -p 5000:5000 registry
[root@test ~]# docker tag redis 192.168.0.35:5000/redis:latest
[root@test ~]# docker push 192.168.0.35:5000/redis:latest

部署redis节点的docker 使用ansible部署

[root@test ~]# vim /etc/hosts
192.168.0.71 redis1
192.168.0.72 redis2
192.168.0.73 redis3
[root@lianxi-0001 ~]# yum -y install ansible
[root@lianxi-0001 ~]# vim /etc/ansible/ansible.cfg
[defaults]
inventory      = /etc/ansible/hosts
host_key_checking = False
[root@lianxi-0001 ~]# vim /etc/ansible/hosts
[redis]
redis[1:3]
[root@test ~]# ansible redis --list-hosts        #提前先部署秘钥
  hosts (3):
    redis1
    redis2
    redis3
[root@test ~]# ansible redis -m shell -a 'yum -y install docker-engine;systemctl start docker'
[root@test ~]# ansible redis -m copy -a 'src=/etc/docker/daemon.json dest=/etc/docker/'
[root@test ~]# ansible redis -m shell -a 'systemctl restart docker;
docker pull 192.168.0.35:5000/redis:latest;
docker tag 192.168.0.35:5000/redis:latest redis'
[root@test ~]# ansible redis -m shell -a 'docker images'    #检查每个节点是否有redis镜像

redis部署

创建redis容器挂载目录

[root@test ~]# ansible redis -m shell -a 'mkdir -p /var/redis/63{79,80}/{conf,logs,data}'

创建redis模板配置文件

[root@test ~]# vim redis.conf.tem
protected-mode no
bind 0.0.0.0                     
port ${PORT}                         
daemonize no                         #关闭以守护进程运行
dir /var/lib/redis
logfile /var/log/redis/redis.log
appendonly yes
appendfilename "appendonly.aof"
appendfsync everysec
cluster-enabled yes
cluster-config-file nodes.conf
cluster-node-timeout 5000
cluster-announce-ip ${IP}           #指定redis集群ip,即为宿主机ip
cluster-announce-port ${PORT}
cluster-announce-bus-port 1${PORT}

使用envsubst命令修改模板文件后分发到各个节点(也可以使用ansible-playbook完成)

[root@test ~]# for ip in 192.168.0.7{1..3}
do 
for port in 63{79,80} 
do 
scp redis.conf.tem $ip: 
ssh $ip "export IP=$ip PORT=$port;envsubst < redis.conf.tem > /var/redis/${port}/conf/redis.conf"
done
done

在节点上检查结构

[root@redis1 ~]# tree /var/redis/
/var/redis/
├── 6379
│   ├── conf
│   │   └── redis.conf
│   ├── data
│   └── logs
│      
└── 6380
    ├── conf
    │   └── redis.conf
    ├── data
    └── logs

检查redis.conf文件,如查看redis1上的配置文件

[root@test ~]# ansible redis1 -m shell -a 'cat /var/redis/6379/conf/redis.conf'
protected-mode no
bind 192.168.0.0          
port 6379
daemonize no
dir /var/lib/redis
logfile /var/log/redis/redis.log
appendonly yes
appendfilename "appendonly.aof"
appendfsync everysec
cluster-enabled yes
cluster-config-file nodes.conf
cluster-node-timeout 5000
cluster-announce-ip 192.168.0.71
cluster-announce-port 6379
cluster-announce-bus-port 16379

创建redis实例
-p 端口映射,需要把每个实例的63(79|80) 和163(79|80)端口都映射出来
-v 指定目录映射
–name 给容器命名
-d 容器放入后台

[root@test ~]# for i in 63{79,80}
do  
ansible redis -m shell -a "docker run -itd -p $i:$i -p 1$i:1$i -v /var/redis/$i/logs:/var/log/redis -v /var/redis/$i/data:/var/lib/redis/ -v /var/redis/$i/conf/redis.conf:/etc/redis/redis.conf --name redis$i redis" 
done

验证redis实例随便进入一台redis服务器,进入redis容器内部,执行完命令后看见6个PONG则验证成功

[root@test ~]# ssh redis1
[root@redis1 ~]# docker exec -it redis6379 bash
[root@857e857386ee redis-4.0.8]# for ip in 192.168.0.7{1..3}
do 
for port in 63{79,80}
do
redis-cli -h $ip -p $port <<eof
ping
eof
done
done

PONG
PONG
PONG
PONG
PONG
PONG

搭建redis集群

安装ruby环境(也可使用docker里的ruby镜像)
创建3个主redis节点分别为每个redis服务器上的6379端口

[root@test ~]# yum -y install ruby rubygems ruby-devel
[root@test ~]# gem install redis-3.2.1.gem
[root@test ~]# cp workdir/redis-4.0.8/src/redis-trib.rb /sbin/
[root@test ~]# redis-trib.rb create 192.168.0.7{1..3}:6379   
>>> Creating cluster
>>> Performing hash slots allocation on 3 nodes...
Using 3 masters:
192.168.0.71:6379
192.168.0.72:6379
192.168.0.73:6379
M: f2db485427801702c2c6b0fab6c161f7dcbd2c57 192.168.0.71:6379
   slots:0-5460 (5461 slots) master
M: 53cde4277d80fb07cf9138787ba01f4f1128470f 192.168.0.72:6379
   slots:5461-10922 (5462 slots) master
M: d2760a19b13d774e8325e529201e568814c8f5ed 192.168.0.73:6379
   slots:10923-16383 (5461 slots) master
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 192.168.0.71:6379)
M: f2db485427801702c2c6b0fab6c161f7dcbd2c57 192.168.0.71:6379
   slots:0-5460 (5461 slots) master
   0 additional replica(s)
M: 53cde4277d80fb07cf9138787ba01f4f1128470f 192.168.0.72:6379
   slots:5461-10922 (5462 slots) master
   0 additional replica(s)
M: d2760a19b13d774e8325e529201e568814c8f5ed 192.168.0.73:6379
   slots:10923-16383 (5461 slots) master
   0 additional replica(s)
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

为每个主节点添加在不同redis服务器的从节点
格式为

[root@test ~]# redis-trib.rb add-node --slave --master-id f2db485427801702c2c6b0fab6c161f7dcbd2c57 192.168.0.72:6380 192.168.0.71:6379
[root@test ~]# redis-trib.rb add-node --slave --master-id 53cde4277d80fb07cf9138787ba01f4f1128470f 192.168.0.73:6380 192.168.0.71:6379
[root@test ~]# redis-trib.rb add-node --slave --master-id d2760a19b13d774e8325e529201e568814c8f5ed 192.168.0.71:6380 192.168.0.71:6379

检查集群状态

[root@test ~]# redis-trib.rb check 192.168.0.71:6379
>>> Performing Cluster Check (using node 192.168.0.71:6379)
M: f2db485427801702c2c6b0fab6c161f7dcbd2c57 192.168.0.71:6379
   slots:0-5460 (5461 slots) master
   1 additional replica(s)
S: 6e8dc744a314a142e1f3494e304e94b3bd112c13 192.168.0.73:6380
   slots: (0 slots) slave
   replicates 53cde4277d80fb07cf9138787ba01f4f1128470f
S: 2c0abf5fc5e9bc11ae3a4919cb2e9d0377f6d43b 192.168.0.71:6380
   slots: (0 slots) slave
   replicates d2760a19b13d774e8325e529201e568814c8f5ed
M: 53cde4277d80fb07cf9138787ba01f4f1128470f 192.168.0.72:6379
   slots:5461-10922 (5462 slots) master
   1 additional replica(s)
S: bd20aea7765ebb6283476e68c5a2ecd65ddbc47d 192.168.0.72:6380
   slots: (0 slots) slave
   replicates f2db485427801702c2c6b0fab6c161f7dcbd2c57
M: d2760a19b13d774e8325e529201e568814c8f5ed 192.168.0.73:6379
   slots:10923-16383 (5461 slots) master
   1 additional replica(s)
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

[root@test ~]# redis-cli  -c -h 192.168.0.71  -p 6379 <<eof
> cluster info
> eof
cluster_state:ok
cluster_slots_assigned:16384
cluster_slots_ok:16384
cluster_slots_pfail:0
cluster_slots_fail:0
cluster_known_nodes:6
cluster_size:3
cluster_current_epoch:3
cluster_my_epoch:1
cluster_stats_messages_ping_sent:1158
cluster_stats_messages_pong_sent:1156
cluster_stats_messages_sent:2314
cluster_stats_messages_ping_received:1151
cluster_stats_messages_pong_received:1158
cluster_stats_messages_meet_received:5
cluster_stats_messages_received:2314