Redis 集群配置

⾼可⽤：保证redis⼀直处于可⽤状态，即时出现了故障也有备⽤⽅案保证可⽤性

⾼并发：⼀个redis实例已经可以⽀持多达11w并发读操作或者8.1w并发写操作；但是如果对于有更⾼并发需

求的应⽤来说，我们可以通过 读写分离 、 集群配置 来解决⾼并发问题

Redis集群

Redis集群中每个节点是对等的，⽆中⼼结构

数据按照slots分布式存储在不同的redis节点上，节点中的数据可共享，可以动态调整数据的分布

可扩展性强，可以动态增删节点，最多可扩展⾄1000+节点

集群每个节点通过主备（哨兵模式）可以保证其⾼可⽤性

集群搭建

[root@VM-4-3-centos cluster-conf]# cd /usr/local/redis-5.0.5

[root@VM-4-3-centos cluster-conf]# mkdir cluster-conf

[root@VM-4-3-centos cluster-conf]# cat redis.conf | grep -v "#"|grep -v "^$" > cluster-conf/redis-7001.conf

[root@VM-4-3-centos cluster-conf]# cd cluster-conf/

[root@VM-4-3-centos cluster-conf]# ls

redis-7001.conf

[root@VM-4-3-centos cluster-conf]# vim redis-7001.conf

拷⻉6个⽂件，端⼝分别为 7001-7006

[root@VM-4-3-centos cluster-conf]# sed 's/7001/7002/g' redis-7001.conf > redis-7002.conf

[root@VM-4-3-centos cluster-conf]# sed 's/7001/7003/g' redis-7001.conf > redis-7003.conf

[root@VM-4-3-centos cluster-conf]# sed 's/7001/7004/g' redis-7001.conf > redis-7004.conf

[root@VM-4-3-centos cluster-conf]# sed 's/7001/7005/g' redis-7001.conf > redis-7005.conf

[root@VM-4-3-centos cluster-conf]# sed 's/7001/7006/g' redis-7001.conf > redis-7006.conf

启动6个redis实例

[root@VM-4-3-centos cluster-conf]# redis-server redis-7001.conf &

[root@VM-4-3-centos cluster-conf]# redis-server redis-7002.conf &

[root@VM-4-3-centos cluster-conf]# redis-server redis-7003.conf &

[root@VM-4-3-centos cluster-conf]# redis-server redis-7004.conf &

[root@VM-4-3-centos cluster-conf]# redis-server redis-7005.conf &

[root@VM-4-3-centos cluster-conf]# redis-server redis-7006.conf &

查看6个实例是否启动

[root@VM-4-3-centos cluster-conf]# ps -ef|grep redis

root 4789 1 0 10:20 ? 00:00:00 redis-server *:7001 [cluster]

root 4794 1 0 10:20 ? 00:00:00 redis-server *:7002 [cluster]

root 4799 1 0 10:20 ? 00:00:00 redis-server *:7003 [cluster]

root 4806 1 0 10:21 ? 00:00:00 redis-server *:7004 [cluster]

root 4811 1 0 10:21 ? 00:00:00 redis-server *:7005 [cluster]

root 4816 1 0 10:21 ? 00:00:00 redis-server *:7006 [cluster]

启动集群

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster create 81.68.82.130:7001 81.68.82.130:7002 81.68.82.130:7003 81.68.82.130:7004   81.68.82.130:7005 81.68.82.130:7006 --cluster-replicas 1

连接集群：

[root@VM-4-3-centos cluster-conf]# redis-cli -p 7001 -c

集群管理

如果集群启动失败：等待节点加⼊

1. 云服务器检查安全组是否放⾏redis实例端⼝，以及+10000的端⼝

2. Linux防⽕墙是否放⾏redis服务（关闭防⽕墙）

3. Linux状态（top）---- 更换云主机操作系统

4. redis配置⽂件错误

创建集群：

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster create 81.68.82.130:7001 81.68.82.130:7002 81.68.82.130:7003 81.68.82.130:7004 81.68.82.130:7005 81.68.82.130:7006 --cluster-replicas 1

查看集群状态

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster info 81.68.82.130:7001

47.96.11.185:7001 (4678478a...) -> 2 keys | 5461 slots | 1 slaves.

47.96.11.185:7002 (e26eaf2a...) -> 0 keys | 5462 slots | 1 slaves.

47.96.11.185:7003 (5752eb20...) -> 1 keys | 5461 slots | 1 slaves.

[OK] 3 keys in 3 masters.

0.00 keys per slot on average.

平衡节点的数据槽数

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster rebalance 81.68.82.130:7001

>>> Performing Cluster Check (using node 81.68.82.130:7001)

[OK] All nodes agree about slots configuration.

>>> Check for open slots...

>>> Check slots coverage...

[OK] All 16384 slots covered.

*** No rebalancing needed! All nodes are within the 2.00% threshold.

迁移节点槽

以下是将 7001端口 的 5459个节点槽迁移到 7002端口上

删除节点

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster del-node 81.68.82.130:7001 2bbb2237941d3cbc51c841bf9b4bd6c6e9db94a3

>>> Removing node 2bbb2237941d3cbc51c841bf9b4bd6c6e9db94a3 from cluster 81.68.82.130:7001

>>> Sending CLUSTER FORGET messages to the cluster...

>>> SHUTDOWN the node.

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster info 81.68.82.130:7002

81.68.82.130:7002 (a5cd5c1f...) -> 2 keys | 10923 slots | 2 slaves.

81.68.82.130:7003 (609295d3...) -> 0 keys | 5461 slots | 1 slaves.

[OK] 2 keys in 2 masters.

0.00 keys per slot on average.

添加节点

[root@VM-4-3-centos cluster-conf]# redis-cli --cluster add-node 81.68.82.130:7007 81.68.82.130:7002

>>> Adding node 81.68.82.130:7007 to cluster 81.68.82.130:7002

>>> Performing Cluster Check (using node 81.68.82.130:7002)

M: a5cd5c1f696d367aac2ab511c20401c5ca693df8 81.68.82.130:7002

  slots:[0-10922] (10923 slots) master

  2 additional replica(s)

S: 55c3cbd57c90a73cac4dd06872ad7fa837fcc5ed 81.68.82.130:7005

  slots: (0 slots) slave

  replicates a5cd5c1f696d367aac2ab511c20401c5ca693df8

S: a8d7c675745dcfa08f183c66e0c8870fea234180 81.68.82.130:7006

  slots: (0 slots) slave

  replicates a5cd5c1f696d367aac2ab511c20401c5ca693df8

M: 609295d3d82f46cc77775e0a122733067e87bce2 81.68.82.130:7003

  slots:[10923-16383] (5461 slots) master

  1 additional replica(s)

S: c291dddb5fc9f5a9a8da5a22988030b26caa31b4 81.68.82.130:7004

  slots: (0 slots) slave

  replicates 609295d3d82f46cc77775e0a122733067e87bce2

[OK] All nodes agree about slots configuration.

>>> Check for open slots...

>>> Check slots coverage...

[OK] All 16384 slots covered.

>>> Send CLUSTER MEET to node 81.68.82.130:7007 to make it join the cluster.

[OK] New node added correctly.