基于Redis6.2.6版本部署Redis Cluster集群1.Redis6.2.6简介以及环境规划
在Redis6.x版本中主要增加了多线程的新特性，多线性对于高并发场景是非常有必要的。
Redis6.x新特性如下：
多线程IO重新设计了客户端缓存功能RESP3协议支持SSLACL权限控制提升了RDB日志加载速度发布官方的Redis集群代理模块 Redis Cluster proxy
Redis Cluster集群原理可以查看之前发布的文章。
环境规划：
IP主机名端口号节点
192.168.20.10redis-cluster6701master192.168.20.10redis-cluster6702slave—>6705的master192.168.20.10redis-cluster6703master192.168.20.10redis-cluster6704slave—>6701的master192.168.20.10redis-cluster6705master192.168.20.10redis-cluster6706slave—>6703的master架构图：
Reids集群采用三主三从交叉复制架构，由于服务器数量有限，在一台机器中模拟出集群的效果，在实际生产环境中，需要准备三台机器，每台机器中分别部署两台Redis节点，一主一从，交叉备份。
2.二进制安装Redis程序2.1.二进制安装redis6.2.61.下载reids[root@k8s-master1 /data/]# wget https://download.redis.io/releases/redis-6.2.6.tar.gz2.解压并安装redis[root@k8s-master1 /data/]# tar xf redis-6.2.6.tar.gz[root@k8s-master1 /data/]# cd redis-6.2.6[root@k8s-master1 /data/redis-6.2.6]# make2.2.创建Reids Cluster集群目录1.创建Redis Cluster各集群节点的配置文件存放路径[root@k8s-master ~]# mkdir /data/redis-6.2.6-cluster/{bin,conf,data,var,logs} -p[root@k8s-master ~]# ll /data/redis-6.2.6-cluster/总用量 0drwxr-xr-x 2 root root 6 3月 31 15:36 bin#可执行文件drwxr-xr-x 2 root root 6 3月 31 15:36 conf#配置文件drwxr-xr-x 2 root root 6 3月 31 15:36 data#数据文件drwxr-xr-x 2 root root 6 3月 31 15:36 log#日志文件drwxr-xr-x 2 root root 6 3月 31 15:36 var#pid文件2.准备redis可执行命令[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-benchmark /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-check-aof /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-check-rdb /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-cli /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-sentinel /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-server /data/redis-6.2.6-cluster/bin[root@k8s-master1 /data/redis-6.2.6]# cp src/redis-trib.rb /data/redis-6.2.6-cluster/bin3.配置Redis Cluster三主三从交叉复制集群3.1.准备六个节点的redis配置文件
redis6.2.6版本配置文件所有内容如下，主要配置带注释的行。
bind 0.0.0.0protected-mode no#关闭保护模式port 6701#端口号tcp-backlog 511timeout 0tcp-keepalive 300daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6701.pid#pid存放loglevel noticelogfile /data/redis-6.2.6-cluster/logs/redis_6701.log#日志存放路径databases 16always-show-logo yes#是否显示总日志set-proc-title yesproc-title-template "{title} {listen-addr} {server-mode}"stop-writes-on-bgsave-error yesrdbcompression yesrdbchecksum yesdbfilename redis_6701.rdb#持久化数据文件名称rdb-del-sync-files nodir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径replica-serve-stale-data yesreplica-read-only yesrepl-diskless-sync norepl-diskless-sync-delay 5repl-diskless-load disabledrepl-disable-tcp-nodelay noreplica-priority 100acllog-max-len 128lazyfree-lazy-eviction nolazyfree-lazy-expire nolazyfree-lazy-server-del noreplica-lazy-flush nolazyfree-lazy-user-del nolazyfree-lazy-user-flush nooom-score-adj nooom-score-adj-values 0 200 800disable-thp yesappendonly noappendfilename "appendonly.aof"appendfsync everysecno-appendfsync-on-rewrite noauto-aof-rewrite-percentage 100auto-aof-rewrite-min-size 64mbaof-load-truncated yesaof-use-rdb-preamble yeslua-time-limit 5000cluster-enabled yes#开启集群模式cluster-config-file nodes_6701.conf #集群模式配置文件名称cluster-node-timeout 15000#集群超时时间slowlog-log-slower-than 10000slowlog-max-len 128latency-monitor-threshold 0notify-keyspace-events ""hash-max-ziplist-entries 512hash-max-ziplist-value 64list-max-ziplist-size -2list-compress-depth 0set-max-intset-entries 512zset-max-ziplist-entries 128zset-max-ziplist-value 64hll-sparse-max-bytes 3000stream-node-max-bytes 4096stream-node-max-entries 100activerehashing yesclient-output-buffer-limit normal 0 0 0client-output-buffer-limit replica 256mb 64mb 60client-output-buffer-limit pubsub 32mb 8mb 60hz 10dynamic-hz yesaof-rewrite-incremental-fsync yesrdb-save-incremental-fsync yesjemalloc-bg-thread yes
1.复制二进制中的配置文件
[root@k8s-master redis-6.2.6]# cp redis.conf /data/redis-6.2.6-cluster/conf/
2.Reids 6701节点配置文件内容
[root@k8s-master conf]# vim redis_6701.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6701#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6701.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6701.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6701.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6701.conf #集群模式配置文件名称cluster-node-timeout 15000#集群超时时间
3.Reids 6702节点配置文件内容
[root@k8s-master conf]# vim redis_6702.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6702#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6702.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6702.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6702.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6702.conf #集群模式配置文件名称cluster-node-timeout 15000#集群超时时间
4.Reids 6703节点配置文件内容
[root@k8s-master conf]# vim redis_6702.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6703#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6703.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6703.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6703.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6703.conf #集群模式配置文件名称cluster-node-timeout 15000
5.Reids 6704节点配置文件内容
[root@k8s-master conf]# vim redis_6702.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6704#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6704.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6704.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6704.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6704.conf #集群模式配置文件名称cluster-node-timeout 15000
6.Reids 6705节点配置文件内容
[root@k8s-master conf]# vim redis_6702.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6705#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6705.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6705.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6705.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6705.conf #集群模式配置文件名称cluster-node-timeout 15000
7.Reids 6706节点配置文件内容
[root@k8s-master conf]# vim redis_6702.confbind 0.0.0.0#监听地址protected-mode no#关闭保护模式port 6706#端口号daemonize yes#后台运行pidfile /data/redis-6.2.6-cluster/var/redis_6706.pid#pid存放路径logfile /data/redis-6.2.6-cluster/logs/redis_6706.log#日志存放路径always-show-logo yes #是否显示总日志dbfilename redis_6706.rdb#持久化数据文件名称dir /data/redis-6.2.6-cluster/data#持久化数据文件存放路径cluster-enabled yes#开启集群模式cluster-config-file nodes_6706.conf #集群模式配置文件名称cluster-node-timeout 150003.2.将六个节点全部启动
1）启动redis各个节点
[root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6701.conf [root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6702.conf [root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6703.conf [root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6704.conf [root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6705.conf [root@k8s-master ~]# /data/redis-6.2.6-cluster/bin/redis-server /data/redis-6.2.6-cluster/conf/redis_6706.conf
2）查看启动的进程
[root@k8s-master ~]# ps aux | grep redisroot 22166 0.2 0.0 164964 3208 ? Ssl 16:17 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6701 [cluster]root 22179 0.2 0.0 164964 3212 ? Ssl 16:17 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6702 [cluster]root 22235 0.2 0.0 164964 3216 ? Ssl 16:17 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6703 [cluster]root 23748 0.1 0.0 164964 3220 ? Ssl 16:18 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6704 [cluster]root 24093 0.1 0.0 164964 3224 ? Ssl 16:19 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6705 [cluster]root 24513 0.0 0.0 164964 3212 ? Ssl 16:19 0:00 /data/redis-6.2.6-cluster/bin/redis-server 0.0.0.0:6706 [cluster]
3）查看各个节点生成的文件
3.3.配置集群节点之间相互发现
在任意一个节点中操作即可。
[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6701127.0.0.1:6701> CLUSTER MEET 192.168.20.10 6702OK127.0.0.1:6701> CLUSTER MEET 192.168.20.10 6703OK127.0.0.1:6701> CLUSTER MEET 192.168.20.10 6704OK127.0.0.1:6701> CLUSTER MEET 192.168.20.10 6705OK127.0.0.1:6701> CLUSTER MEET 192.168.20.10 6706OK127.0.0.1:6701> CLUSTER NODES1de9d406a489e3aedc2f7743114f21700de5d151 192.168.20.10:6702@16702 master – 0 1648715495901 1 connectedf1d6f55a4a66fb97b3be2dff70cf0b8984a9de42 192.168.20.10:6704@16704 master – 0 1648715494000 0 connected14bd716bdc2726bfbf39eb14c88f072970059676 192.168.20.10:6706@16706 master – 0 1648715494894 5 connected7eb4466a8558ab389a9d92d115d5b420fa2a8ffc 192.168.20.10:6701@16701 myself,master – 0 1648715494000 2 connectedb7e93e3f3925c6db664afc59cb08c2abced2933d 192.168.20.10:6703@16703 master – 0 1648715494000 3 connectedc502c900224a0ae6173631746de92d012bd43da3 192.168.20.10:6705@16705 master – 0 1648715494000 4 connected
操作完成后会同步各节点的集群配置文件中。
3.4.为集群中的充当Master的节点分配槽位
1）分配槽位
[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6701 cluster addslots {0..5461}OK[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6703 cluster addslots {5462..10922}OK[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6705 cluster addslots {10923..16383}OK
2）查看集群状态
分配完槽位之后集群的状态就会处于ok。
[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6701127.0.0.1:6701> CLUSTER INFOcluster_state:okcluster_slots_assigned:16384cluster_slots_ok:16384cluster_slots_pfail:0cluster_slots_fail:0cluster_known_nodes:6cluster_size:3cluster_current_epoch:5cluster_my_epoch:2cluster_stats_messages_ping_sent:248cluster_stats_messages_pong_sent:246cluster_stats_messages_meet_sent:5cluster_stats_messages_sent:499cluster_stats_messages_ping_received:246cluster_stats_messages_pong_received:253cluster_stats_messages_received:499
3）查看集群中的节点信息
虽然现在槽位已经分配完毕，但是当下集群中所有的节点都处于Master角色，还是无法使用。
127.0.0.1:6701> CLUSTER NODES1de9d406a489e3aedc2f7743114f21700de5d151 192.168.20.10:6702@16702 master – 0 1648715802089 1 connectedf1d6f55a4a66fb97b3be2dff70cf0b8984a9de42 192.168.20.10:6704@16704 master – 0 1648715802000 0 connected14bd716bdc2726bfbf39eb14c88f072970059676 192.168.20.10:6706@16706 master – 0 1648715802000 5 connected7eb4466a8558ab389a9d92d115d5b420fa2a8ffc 192.168.20.10:6701@16701 myself,master – 0 1648715803000 2 connected 0-5461b7e93e3f3925c6db664afc59cb08c2abced2933d 192.168.20.10:6703@16703 master – 0 1648715803095 3 connected 5462-10922c502c900224a0ae6173631746de92d012bd43da3 192.168.20.10:6705@16705 master – 0 1648715804100 4 connected 10923-16383
接下来去配置三主三从交叉复制模式，让没有槽位的节点复制有槽位的节点，也就是主从关系。
3.5.配置三主三从交叉复制模式
1）获取集群主节点的信息
[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6701 cluster nodes | egrep ‘6701|6703|6705’ | awk ‘{print $1,$2}’7eb4466a8558ab389a9d92d115d5b420fa2a8ffc 192.168.20.10:6701@16701b7e93e3f3925c6db664afc59cb08c2abced2933d 192.168.20.10:6703@16703c502c900224a0ae6173631746de92d012bd43da3 192.168.20.10:6705@16705
2）配置交叉复制
1.配置6702节点复制6705节点，充当6705 master节点的slave[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6702127.0.0.1:6702> CLUSTER REPLICATE c502c900224a0ae6173631746de92d012bd43da3OK2.配置6704节点复制6701节点，充当6701 master节点的slave[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6704127.0.0.1:6704> CLUSTER REPLICATE 7eb4466a8558ab389a9d92d115d5b420fa2a8ffcOK3.配置6706节点复制6703节点，充当6703 master节点的slave[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6706127.0.0.1:6706> CLUSTER REPLICATE b7e93e3f3925c6db664afc59cb08c2abced2933dOK
3）查看集群节点信息
已经是三主三从交叉复制集群模式了。
[root@redis-cluster ~]# /data/redis-6.2.6-cluster/bin/redis-cli -p 6701127.0.0.1:6701> CLUSTER NODES1de9d406a489e3aedc2f7743114f21700de5d151 192.168.20.10:6702@16702 slave c502c900224a0ae6173631746de92d012bd43da3 0 1648716239215 4 connectedf1d6f55a4a66fb97b3be2dff70cf0b8984a9de42 192.168.20.10:6704@16704 slave 7eb4466a8558ab389a9d92d115d5b420fa2a8ffc 0 1648716236175 2 connected14bd716bdc2726bfbf39eb14c88f072970059676 192.168.20.10:6706@16706 slave b7e93e3f3925c6db664afc59cb08c2abced2933d 0 1648716238000 3 connected7eb4466a8558ab389a9d92d115d5b420fa2a8ffc 192.168.20.10:6701@16701 myself,master – 0 1648716237000 2 connected 0-5461b7e93e3f3925c6db664afc59cb08c2abced2933d 192.168.20.10:6703@16703 master – 0 1648716238192 3 connected 5462-10922c502c900224a0ae6173631746de92d012bd43da3 192.168.20.10:6705@16705 master – 0 1648716237184 4 connected 10923-163834.快速搭建Redis Cluster集群
前提条件需要先将6个节点搭建起来，再使用下面的命令快速将三主三从集群部署起来。
[root@redis-cluster redis-6.2.6-cluster]# /data/redis-6.2.6-cluster/bin/redis-cli –cluster create 192.168.20.10:6701 192.168.20.10:6702 192.168.20.10:6703 192.168.20.10:6704 192.168.20.10:6705 192.168.20.10:6706 –cluster-replicas 1 Can I set the above configuration? (type ‘yes’ to accept): yes#–cluster起到的作用就是省掉了节点发现、分配槽位、交叉复制的过程，自动替我们分配。
从输出信息中我们清晰的看到主从的复制关系以及分配的槽位。
>>> Performing hash slots allocation on 6 nodes…Master[0] -> Slots 0 – 5460Master[1] -> Slots 5461 – 10922Master[2] -> Slots 10923 – 16383Adding replica 192.168.20.10:6705 to 192.168.20.10:6701Adding replica 192.168.20.10:6706 to 192.168.20.10:6702Adding replica 192.168.20.10:6704 to 192.168.20.10:6703
6701、6702、6703节点为主节点，6704、6705、6706节点为从节点。
6704复制6703的主节点、6705复制6701的主节点、6706复制6702的主节点。
架构图如下：
我们可以失望，但不能盲目。
