本文是Redis集群学习的实践总结(基于Redis 6.0+),详细介绍逐步搭建Redis集群环境的过程,并完成集群伸缩的实践。

Redis集群简介

Redis集群(Redis Cluster) 是Redis提供的分布式数据库方案,通过 分片(sharding) 来进行数据共享,并提供复制和故障转移功能。相比于主从复制、哨兵模式,Redis集群实现了较为完善的高可用方案,解决了存储能力受到单机限制,写操作无法负载均衡的问题。

本文是Redis集群学习的实践总结,详细介绍逐步搭建Redis集群环境的过程,并完成集群伸缩的实践。

发布过很多 redis 的优质文章,可以关注微信公众号 Java后端,回复 666 下载这一本 Java技术栈文档。

1 Redis集群环境搭建

方便起见,这里集群环境的所有节点全部位于同一个服务器上,共6个节点以端口号区分,3个主节点+3个从节点。集群的简单架构如图:



helm redis集群安装 redis6.0集群搭建_java

本文基于最新的Redis 6.0+,直接从github下载最新的源码编译获得常用工具 redis-server ,  redis-cli 。值得注意的是,从Redis 5.0以后的版本,集群管理软件 redis-trib.rb 被集成到 redis-cli 客户端工具中(详细可参考cluster-tutorial)。

本节介绍集群环境搭建时,并未借助 redis-trib.rb 快速管理,而是按照标准步骤一步步搭建,这也是为了熟悉集群管理的基本步骤。在集群伸缩实践一节将借助 redis-trib.rb 完成集群重新分片工作。

集群的搭建可以分为四步:

  1. 启动节点:将节点以集群方式启动,此时节点是独立的。
  2. 节点握手:将独立的节点连成网络。
  3. 槽指派:将16384个槽位分配给主节点,以达到分片保存数据库键值对的效果。
  4. 主从复制:为从节点指定主节点。
1.1 启动节点

每个节点初始状态仍为 Master服务器,唯一不同的是:使用 Cluster 模式启动。需要对配置文件进行修改,以端口号为6379的节点为例,主要修改如下几项:

# redis_6379_cluster.conf
port 6379
cluster-enabled yes
cluster-config-file "node-6379.conf"
logfile "redis-server-6379.log"
dbfilename "dump-6379.rdb"
daemonize yes

其中 cluster-config-file 参数指定了集群配置文件的位置,每个节点在运行过程中,会维护一份集群配置文件;每当集群信息发生变化时(如增减节点),集群内所有节点会将最新信息更新到该配置文件;当节点重启后,会重新读取该配置文件,获取集群信息,可以方便的重新加入到集群中。也就是说,当Redis节点以集群模式启动时,会首先寻找是否有集群配置文件,如果有则使用文件中的配置启动,如果没有,则初始化配置并将配置保存到文件中。集群配置文件由Redis节点维护,不需要人工修改。

为6个节点修改好相应的配置文件后,即可利用 redis-server redis_xxxx_cluster.conf 工具启动6个服务器(xxxx表示端口号,对应相应的配置文件)。利用ps命令查看进程:

$ ps -aux | grep redis
... 800  0.1  0.0  49584  2444 ?        Ssl  20:42   0:00 redis-server 127.0.0.1:6379 [cluster]
... 805  0.1  0.0  49584  2440 ?        Ssl  20:42   0:00 redis-server 127.0.0.1:6380 [cluster]
... 812  0.3  0.0  49584  2436 ?        Ssl  20:42   0:00 redis-server 127.0.0.1:6381 [cluster]
... 817  0.1  0.0  49584  2432 ?        Ssl  20:43   0:00 redis-server 127.0.0.1:6479 [cluster]
... 822  0.0  0.0  49584  2380 ?        Ssl  20:43   0:00 redis-server 127.0.0.1:6480 [cluster]
... 827  0.5  0.0  49584  2380 ?        Ssl  20:43   0:00 redis-server 127.0.0.1:6481 [cluster]
1.2 节点握手

1.1中将每个节点启动后,节点间是相互独立的,他们都处于一个只包含自己的集群当中,以端口号6379的服务器为例,利用 CLUSTER NODES 查看当前集群包含的节点。

127.0.0.1:6379> CLUSTER NODES
37784b3605ad216fa93e976979c43def42bf763d :6379@16379 myself,master - 0 0 0 connected 449 4576 5798 7568 8455 12706

我们需要将各个独立的节点连接起来,构成一个包含多个节点的集群,使用 CLUSTER MEET 命令。

$ redis-cli -p 6379 -c          # -c 选项指定以Cluster模式运行redis-cli 127.0.0.1:6379> CLUSTER MEET 127.0.0.1 6380 OK 127.0.0.1:6379> CLUSTER MEET 127.0.0.1 6381 OK 127.0.0.1:6379> CLUSTER MEET 127.0.0.1 6480 OK 127.0.0.1:6379> CLUSTER MEET 127.0.0.1 6381 OK 127.0.0.1:6379> CLUSTER MEET 127.0.0.1 6382 OK

再次查看此时集群中包含的节点情况:

127.0.0.1:6379> CLUSTER NODES
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603632309283 4 connected
87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,master - 0 1603632308000 1 connected
51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603632310292 2 connected
9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 master - 0 1603632309000 5 connected
4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603632308000 3 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 master - 0 1603632311302 0 connected

可以发现此时6个节点均作为主节点加入到集群中, CLUSTER NODES 返回的结果各项含义如下:

<id> <ip:port@cport> <flags> <master> <ping-sent> <pong-recv> <config-epoch> <link-state> <slot> <slot> ... <slot>
  • 节点id: 由40个16进制字符串组成,节点id只在集群初始化时创建一次,然后保存到集群配置文件(即前文提到的cluster-config-file)中,以后节点重新启动时会直接在集群配置文件中读取。
  • port@cport: 前者为普通端口,用于为客户端提供服务;后者为集群端口,分配方法为:普通端口+10000,只用于节点间的通讯。

其余各项的详细解释可以参考官方文档cluster nodes。

1.3 槽指派

Redis集群通过分片(sharding)的方式保存数据库的键值对,整个数据库被分为16384个槽(slot),数据库每个键都属于这16384个槽的一个,集群中的每个节点都可以处理0个或者最多16384个slot。

槽是数据管理和迁移的基本单位。当数据库中的16384个槽都分配了节点时,集群处于上线状态(ok);如果有任意一个槽没有分配节点,则集群处于下线状态(fail)。

注意,只有主节点有处理槽的能力,如果将槽指派步骤放在主从复制之后,并且将槽位分配给从节点,那么集群将无法正常工作(处于下线状态)。

利用 CLUSTER ADDSLOTS

redis-cli  -p 6379 cluster addslots {0..5000}
redis-cli  -p 6380 cluster addslots {5001..10000}
redis-cli  -p 6381 cluster addslots {10001..16383}

槽指派后集群中节点情况如下:

127.0.0.1:6379> CLUSTER NODES
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603632880310 4 connected 5001-10000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,master - 0 1603632879000 1 connected 0-5000 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603632879000 2 connected 10001-16383 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 master - 0 1603632878000 5 connected
4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603632880000 3 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 master - 0 1603632881317 0 connected

127.0.0.1:6379> CLUSTER INFO
cluster_state:ok                        # 集群处于上线状态
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:5 cluster_my_epoch:1 cluster_stats_messages_ping_sent:4763 cluster_stats_messages_pong_sent:4939 cluster_stats_messages_meet_sent:5 cluster_stats_messages_sent:9707 cluster_stats_messages_ping_received:4939 cluster_stats_messages_pong_received:4768 cluster_stats_messages_received:9707
1.4 主从复制

上述步骤后,集群节点均作为主节点存在,仍不能实现Redis的高可用,配置主从复制之后,才算真正实现了集群的高可用功能。

CLUSTER REPLICATE <node_id> 用来让集群中接收命令的节点成为 node_id 所指定节点的从节点,并开始对主节点进行复制。

redis-cli  -p 6479 cluster replicate 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52
redis-cli  -p 6480 cluster replicate c47598b25205cc88abe2e5094d5bfd9ea202335f
redis-cli  -p 6481 cluster replicate 51081a64ddb3ccf5432c435a8cf20d45ab795dd8

127.0.0.1:6379> CLUSTER NODES
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603633105211 4 connected 5001-10000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,master - 0 1603633105000 1 connected 0-5000 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603633105000 2 connected 10001-16383 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603633107229 5 connected
4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 slave 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 0 1603633106221 3 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603633104000 4 connected

顺带补充,上述步骤1.2,1.3,1.4可以利用 redis-trib.rb 工具整体实现,在Redis 5.0之后直接利用 redis-cli 完成,参考命令如下:

redis-cli --cluster create  127.0.0.1:6379 127.0.0.1:6479  127.0.0.1:6380 127.0.0.1:6480  127.0.0.1:6381 127.0.0.1:6481  --cluster-replicas 1

--cluster-replicas 1 指示给定的创建节点列表是以主节点+从节点对组成的。

1.5 在集群中执行命令

集群此时处于上线状态,可以通过客户端向集群中的节点发送命令。接收命令的节点会计算出命令要处理的键属于哪个槽,并检查这个槽是否指派给自己。

  • 如果键所在的slot刚好指派给了当前节点,会直接执行这个命令。
  • 否则,节点向客户端返回 MOVED 错误,指引客户端转向 redirect 至正确的节点,并再次发送此前的命令。

此处,我们利用 CLUSTER KEYSLOT 查看到键 name 所在槽号为5798(被分配在6380节点),当对此键操作时,会被重定向到相应的节点。对键 fruits 的操作与此类似。

127.0.0.1:6379> CLUSTER KEYSLOT name
(integer) 5798
127.0.0.1:6379> set name huey -> Redirected to slot [5798] located at 127.0.0.1:6380 OK 127.0.0.1:6380>

127.0.0.1:6379> get fruits -> Redirected to slot [14943] located at 127.0.0.1:6381
"apple"
127.0.0.1:6381>

值得注意的是,当我们将命令通过客户端发送给一个从节点时,命令会被重定向至对应的主节点。

127.0.0.1:6480> KEYS *
1) "name"
127.0.0.1:6480> get name -> Redirected to slot [5798] located at 127.0.0.1:6380
"huey"
1.6 集群故障转移

集群中主节点下线时,复制此主节点的所有的从节点将会选出一个节点作为新的主节点,并完成故障转移。和主从复制的配置相似,当原先的从节点再次上线,它会被作为新主节点的的从节点存在于集群中。

下面模拟6379节点宕机的情况(将其SHUTDOWN),可以观察到其从节点6479将作为新的主节点继续工作。

462:S 26 Oct 14:08:12.750 * FAIL message received from c47598b25205cc88abe2e5094d5bfd9ea202335f about 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 462:S 26 Oct 14:08:12.751 # Cluster state changed: fail 462:S 26 Oct 14:08:12.829 # Start of election delayed for 595 milliseconds (rank #0, offset 9160). 462:S 26 Oct 14:08:13.434 # Starting a failover election for epoch 6. 462:S 26 Oct 14:08:13.446 # Failover election won: I'm the new master.
462:S 26 Oct 14:08:13.447 # configEpoch set to 6 after successful failover 462:M 26 Oct 14:08:13.447 # Setting secondary replication ID to d357886e00341b57bf17e46b6d9f8cf53b7fad21, valid up to offset: 9161. New replication ID is adbf41b16075ea22b17f145186c53c4499864d5b 462:M 26 Oct 14:08:13.447 * Discarding previously cached master state. 462:M 26 Oct 14:08:13.448 # Cluster state changed: ok

6379节点从宕机状态恢复后,将作为6380节点的从节点存在。

127.0.0.1:6379> CLUSTER NODES
51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603692968000 2 connected 10001-16383 c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603692968504 0 connected 5001-10000 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603692967495 6 connected 0-5000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603692964000 1 connected
9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603692967000 4 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603692967000 5 connected

前文提到 cluster-config-file 会记录下集群节点的状态,打开节点6379的配置文件 nodes-6379.conf ,可以看到 CLUSTER NODES 所示信息均被保存在配置文件中:

51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603694920206 2 connected 10001-16383 c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603694916000 0 connected 5001-10000 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603694920000 6 connected 0-5000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603694918000 1 connected
9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603694919000 4 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603694919200 5 connected
vars currentEpoch 6 lastVoteEpoch 0

2 集群伸缩实践

集群伸缩的关键在于对集群的进行重新分片,实现槽位在节点间的迁移。本节将以在集群中添加节点和删除节点为例,对槽迁移进行实践。

借助于 redis-cli 中集成的 redis-trib.rb 工具进行槽位的管理,工具的帮助菜单如下:

$ redis-cli --cluster help
Cluster Manager Commands:
  create         host1:port1 ... hostN:portN --cluster-replicas <arg> check          host:port --cluster-search-multiple-owners
  info           host:port
  fix            host:port --cluster-search-multiple-owners --cluster-fix-with-unreachable-masters
  reshard        host:port --cluster-from <arg>
                 --cluster-to <arg>
                 --cluster-slots <arg>
                 --cluster-yes --cluster-timeout <arg>
                 --cluster-pipeline <arg>
                 --cluster-replace
  rebalance      host:port --cluster-weight <node1=w1...nodeN=wN>
                 --cluster-use-empty-masters --cluster-timeout <arg>
                 --cluster-simulate --cluster-pipeline <arg>
                 --cluster-threshold <arg>
                 --cluster-replace
  add-node       new_host:new_port existing_host:existing_port --cluster-slave --cluster-master-id <arg> del-node       host:port node_id
  call           host:port command arg arg .. arg set-timeout    host:port milliseconds
  import         host:port --cluster-from <arg>
                 --cluster-copy --cluster-replace
  backup         host:port backup_directory
  help
  
For check, fix, reshard, del-node, set-timeout you can specify the host and port of any working node in the cluster.
2.1 集群伸缩-添加节点

考虑在集群中添加两个节点,端口号为6382和6482,其中节点6482对6382进行复制。
(1) 启动节点:按照1.1中介绍的步骤,启动6382和6482节点。

(2) 节点握手:借助 redis-cli --cluster add-node 命令分别添加节点6382和6482。

redis-cli --cluster add-node 127.0.0.1:6382 127.0.0.1:6379 redis-cli --cluster add-node 127.0.0.1:6482 127.0.0.1:6379

$ redis-cli --cluster add-node 127.0.0.1:6382 127.0.0.1:6379
>>> Adding node 127.0.0.1:6382 to cluster 127.0.0.1:6379
>>> Performing Cluster Check (using node 127.0.0.1:6379)
S: 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379 slots: (0 slots) slave
    replicates 4c23b25bd4bcef7f4b77d8287e330ae72e738883
M: 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381 slots:[10001-16383] (6383 slots) master 1 additional replica(s)
M: c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380 slots:[5001-10000] (5000 slots) master 1 additional replica(s)
M: 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479 slots:[0-5000] (5001 slots) master 1 additional replica(s)
S: 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481 slots: (0 slots) slave
    replicates 51081a64ddb3ccf5432c435a8cf20d45ab795dd8
S: 32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480 slots: (0 slots) slave
    replicates c47598b25205cc88abe2e5094d5bfd9ea202335f
[OK] All nodes agree about slots configuration. >>> Check for open slots... >>> Check slots coverage...
[OK] All 16384 slots covered. >>> Send CLUSTER MEET to node 127.0.0.1:6382 to make it join the cluster.
[OK] New node added correctly.
```shell

3) 重新分片:借助 redis-cli --cluster reshard 命令对集群重新分片,使得各节点槽位均衡(分别从节点6379/6380/6381中迁移一些slot到节点6382中)。需要指定:

* 移动的槽位数:最终平均每个主节点有4096个slot,因此总共移动4096 slots
* 接收槽位的目标节点ID:节点6382的ID
* 移出槽位的源节点ID:节点6379/6380/6381的ID

```shell
$ redis-cli --cluster reshard 127.0.0.1 6479
>>> Performing Cluster Check (using node 127.0.0.1:6479)
M: 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479 slots:[0-5000] (5001 slots) master 1 additional replica(s)
S: 32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480 slots: (0 slots) slave
  replicates c47598b25205cc88abe2e5094d5bfd9ea202335f
M: 706f399b248ed3a080cf1d4e43047a79331b714f 127.0.0.1:6482 slots: (0 slots) master
M: af81109fc29f69f9184ce9512c46df476fe693a3 127.0.0.1:6382 slots: (0 slots) master
M: 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381 slots:[10001-16383] (6383 slots) master 1 additional replica(s)
S: 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481 slots: (0 slots) slave
  replicates 51081a64ddb3ccf5432c435a8cf20d45ab795dd8
S: 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379 slots: (0 slots) slave
  replicates 4c23b25bd4bcef7f4b77d8287e330ae72e738883
M: c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380 slots:[5001-10000] (5000 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.
How many slots do you want to move (from 1 to 16384)? 4096 What is the receiving node ID?

(4) 设置主从关系:

redis-cli -p 6482 cluster replicate af81109fc29f69f9184ce9512c46df476fe693a3 

127.0.0.1:6482> CLUSTER NODES
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603694930000 0 connected
51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603694931000 2 connected 11597-16383 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603694932000 2 connected
706f399b248ed3a080cf1d4e43047a79331b714f 127.0.0.1:6482@16482 myself,slave af81109fc29f69f9184ce9512c46df476fe693a3 0 1603694932000 8 connected
87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603694932000 6 connected
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603694933678 0 connected 6251-10000 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603694932669 6 connected 1250-5000 af81109fc29f69f9184ce9512c46df476fe693a3 127.0.0.1:6382@16382 master - 0 1603694933000 9 connected 0-1249 5001-6250 10001-11596
2.2 集群伸缩-删除节点

这里考虑将新添加的两个节点6382和6482删除,需要将节点6382上分配的槽位迁移到其他节点。

(1) 重新分片: 同样借助 redis-cli --cluster reshard 命令,将6382节点上的槽位全部转移到节点6479上。

$ redis-cli --cluster reshard 127.0.0.1 6382
>>> Performing Cluster Check (using node 127.0.0.1:6382)
M: af81109fc29f69f9184ce9512c46df476fe693a3 127.0.0.1:6382 slots:[0-1249],[5001-6250],[10001-11596] (4096 slots) master 1 additional replica(s)
M: 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381 slots:[11597-16383] (4787 slots) master 1 additional replica(s)
S: 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379 slots: (0 slots) slave
    replicates 4c23b25bd4bcef7f4b77d8287e330ae72e738883
S: 32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480 slots: (0 slots) slave
    replicates c47598b25205cc88abe2e5094d5bfd9ea202335f
M: 4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479 slots:[1250-5000] (3751 slots) master 1 additional replica(s)
M: c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380 slots:[6251-10000] (3750 slots) master 1 additional replica(s)
S: 706f399b248ed3a080cf1d4e43047a79331b714f 127.0.0.1:6482 slots: (0 slots) slave
    replicates af81109fc29f69f9184ce9512c46df476fe693a3
S: 9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481 slots: (0 slots) slave
    replicates 51081a64ddb3ccf5432c435a8cf20d45ab795dd8
[OK] All nodes agree about slots configuration. >>> Check for open slots... >>> Check slots coverage...
[OK] All 16384 slots covered.
How many slots do you want to move (from 1 to 16384)? 4096 What is the receiving node ID? 4c23b25bd4bcef7f4b77d8287e330ae72e738883
Please enter all the source node IDs.
Type 'all' to use all the nodes as source nodes for the hash slots.
Type 'done' once you entered all the source nodes IDs.
Source node #1: af81109fc29f69f9184ce9512c46df476fe693a3
Source node #2: done

127.0.0.1:6379> CLUSTER NODES
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603773540922 0 connected 6251-10000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603773539000 1 connected
4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603773541000 10 connected 0-6250 10001-11596 706f399b248ed3a080cf1d4e43047a79331b714f 127.0.0.1:6482@16482 slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603773541000 10 connected
32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603773539000 5 connected
9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603773541931 4 connected
af81109fc29f69f9184ce9512c46df476fe693a3 127.0.0.1:6382@16382 master - 0 1603773539000 9 connected
51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603773540000 2 connected 11597-16383

(2) 删除节点: 利用 redis-cli --cluster del-node 命令依次删除从节点6482和主节点6382。

$ redis-cli --cluster del-node 127.0.0.1:6482 706f399b248ed3a080cf1d4e43047a79331b714f >>> Removing node 706f399b248ed3a080cf1d4e43047a79331b714f from cluster 127.0.0.1:6482
>>> Sending CLUSTER FORGET messages to the cluster... >>> Sending CLUSTER RESET SOFT to the deleted node.
$ redis-cli --cluster del-node 127.0.0.1:6382 af81109fc29f69f9184ce9512c46df476fe693a3 >>> Removing node af81109fc29f69f9184ce9512c46df476fe693a3 from cluster 127.0.0.1:6382
>>> Sending CLUSTER FORGET messages to the cluster... >>> Sending CLUSTER RESET SOFT to the deleted node.

127.0.0.1:6379> CLUSTER NODES
c47598b25205cc88abe2e5094d5bfd9ea202335f 127.0.0.1:6380@16380 master - 0 1603773679121 0 connected 6251-10000 87b7dfacde34b3cf57d5f46ab44fd6fffb2e4f52 127.0.0.1:6379@16379 myself,slave 4c23b25bd4bcef7f4b77d8287e330ae72e738883 0 1603773677000 1 connected
4c23b25bd4bcef7f4b77d8287e330ae72e738883 127.0.0.1:6479@16479 master - 0 1603773678000 10 connected 0-6250 10001-11596 32ed645a9c9d13ca68dba5a147937fb1d05922ee 127.0.0.1:6480@16480 slave c47598b25205cc88abe2e5094d5bfd9ea202335f 0 1603773680130 5 connected
9d587b75bdaed26ca582036ed706df8b2282b0aa 127.0.0.1:6481@16481 slave 51081a64ddb3ccf5432c435a8cf20d45ab795dd8 0 1603773677099 4 connected
51081a64ddb3ccf5432c435a8cf20d45ab795dd8 127.0.0.1:6381@16381 master - 0 1603773678112 2 connected 11597-16383

3 总结

Redis集群环境的搭建主要包括启动节点、节点握手、槽指派和主从复制等四个步骤,集群伸缩同样涉及这几个方面。借助 redis-cli --cluster 命令来管理集群环境,不仅能增加简便性,还能降低操作失误的风险。