Redis三：主从、哨兵和集群的搭建

目录

1 主从搭建

需求

  • 在三台虚拟机上，完成搭建一主两从的主从复制架构(如下图)。要求如下

  1. 从机只能读不能写；

  2. 三台 Redis 的密码均为 123456；

  3. 配置文件、数据文件、日志文件分别存放在 /usr/local/redis-5 目录下的 conf、data、log 文 件夹中(下面哨兵、集群搭建时同理)。

搭建

  • 节点说明

    IP Redis节点
    192.168.192.130:6379 master 节点
    192.168.192.131:6379 slave 节点
    192.168.192.132:6379 slave 节点

  1. 主节点配置文件

    cd /usr/local/redis-5/conf/		# 内容如下

port 6379
daemonize yes
# 日志文件位置
logfile "/usr/local/redis-5/log/redis-6379.log"
# rdb和aof的存放目录
dir "/usr/local/redis-5/data"
dbfilename "dump-6379.rdb"
appendonly yes
appendfilename "appendonly-6379.aof"
# 密码
requirepass 123456
# 放行访问IP限制
bind 0.0.0.0

  2. 从节点配置

    • 各节点需要关闭防火墙(或开放对应端口)，本次使用 systemctl stop firewalld 关闭 防火墙，否则从节点连不上主节点。
    # 将主节点的配置文件 redis-6379.conf 复制到从节点上
scp redis-6379.conf root@192.168.192.131:/usr/local/redis-5/conf/

cd /usr/local/redis-5/conf/
vi redis-6379.conf 			# 内容如下(另一个从节点配置文件与该文件一样,略)

port 6379
daemonize yes
# 日志文件位置
logfile "/usr/local/redis-5/log/redis-6379.log"
# rdb和aof的存放目录
dir "/usr/local/redis-5/data"
dbfilename "dump-6379.rdb"
appendonly yes
appendfilename "appendonly-6379.aof"
# 密码
requirepass 123456
# 放行访问IP限制
bind 0.0.0.0
################### 从节点特有配置 ###############
# 从节点访问主节点密码(必需与主节点 requirepass 一致)
masterauth 123456
# 从节点只读模式
replica-read-only yes
# 从节点IP(新版使用replicaof,老版使用slaveof,两者都行)
slaveof 192.168.192.130 6379

  3. 启动各节点

    # 因为三个节点目录结构一样,故启动命令均一样(记得要先关闭防火墙)
cd /usr/local/redis-5/
bin/redis-server conf/redis-6379.conf

  4. 查看信息

    1. 主节点

      # 连接主节点,执行info replication命令
[root@localhost redis-5]# bin/redis-cli -a 123456 info replication	

role:master				# 表示当前节点是主节点
connected_slaves:2		        # 连接2个从节点,ip、port、state等信息如下
slave0:ip=192.168.192.131,port=6379,state=online,offset=252,lag=0
slave1:ip=192.168.192.132,port=6379,state=online,offset=252,lag=1
master_replid:f06b2e1dcfb28e6648fc8d601413b61be295eca1
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:252
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:252

    2. 从节点

      # 连接任意一个从节点,执行 info replication 命令
bin/redis-cli -a 123456 info replication

role:slave				# 从节点
master_host:192.168.192.130		# 主节点ip
master_port:6379			# 主节点port
master_link_status:up			# 连接状态
master_last_io_seconds_ago:2
master_sync_in_progress:0
slave_repl_offset:602
slave_priority:100
slave_read_only:1
connected_slaves:0
master_replid:f06b2e1dcfb28e6648fc8d601413b61be295eca1
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:602
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:602

  5. 查看日志

    1. 主节点

      cd /usr/local/redis-5/log/
tail -f -n 100 redis-6379.log 

# 开始接受从节点的连接
20090:M 28 Nov 2021 15:48:32.446 * Ready to accept connections
# 接受从节点192.168.192.131:6379的同步
20090:M 28 Nov 2021 15:51:14.998 * Replica 192.168.192.131:6379 asks for synchronization
20090:M 28 Nov 2021 15:51:14.998 * Full resync requested by replica 192.168.192.131:6379
20090:M 28 Nov 2021 15:51:14.998 * Starting BGSAVE for SYNC with target: disk
20090:M 28 Nov 2021 15:51:14.999 * Background saving started by pid 20095
20095:C 28 Nov 2021 15:51:15.008 * DB saved on disk
20095:C 28 Nov 2021 15:51:15.008 * RDB: 4 MB of memory used by copy-on-write
20090:M 28 Nov 2021 15:51:15.020 * Background saving terminated with success
# 192.168.192.131:6379同步成功
20090:M 28 Nov 2021 15:51:15.021 * Synchronization with replica 192.168.192.131:6379 succeeded
# 下面为从节点192.168.192.132:6379的同步信息,与上面相同
20090:M 28 Nov 2021 15:52:32.731 * Replica 192.168.192.132:6379 asks for synchronization
20090:M 28 Nov 2021 15:52:32.731 * Full resync requested by replica 192.168.192.132:6379
20090:M 28 Nov 2021 15:52:32.731 * Starting BGSAVE for SYNC with target: disk
20090:M 28 Nov 2021 15:52:32.732 * Background saving started by pid 20096
20096:C 28 Nov 2021 15:52:32.734 * DB saved on disk
20096:C 28 Nov 2021 15:52:32.734 * RDB: 0 MB of memory used by copy-on-write
20090:M 28 Nov 2021 15:52:32.781 * Background saving terminated with success
20090:M 28 Nov 2021 15:52:32.782 * Synchronization with replica 192.168.192.132:6379 succeeded

    2. 从节点

      # 连接任意一个从节点,查看其日志即可
tail -f -n 100 redis-6379.log 

# 准备接受连接
11648:S 28 Nov 2021 15:51:14.985 * Ready to accept connections
# 连接到主节点192.168.192.130:6379
11648:S 28 Nov 2021 15:51:14.985 * Connecting to MASTER 192.168.192.130:6379
# 主从复制开始
11648:S 28 Nov 2021 15:51:14.985 * MASTER <-> REPLICA sync started
11648:S 28 Nov 2021 15:51:14.986 * Non blocking connect for SYNC fired the event.
11648:S 28 Nov 2021 15:51:14.987 * Master replied to PING, replication can continue...
11648:S 28 Nov 2021 15:51:14.989 * Partial resynchronization not possible (no cached master)
11648:S 28 Nov 2021 15:51:14.999 * Full resync from master: f06b2e1dcfb28e6648fc8d601413b61be295eca1:0
11648:S 28 Nov 2021 15:51:15.013 * MASTER <-> REPLICA sync: receiving 176 bytes from master
11648:S 28 Nov 2021 15:51:15.015 * MASTER <-> REPLICA sync: Flushing old data
11648:S 28 Nov 2021 15:51:15.015 * MASTER <-> REPLICA sync: Loading DB in memory
# 主从复制同步成功
11648:S 28 Nov 2021 15:51:15.015 * MASTER <-> REPLICA sync: Finished with success
11648:S 28 Nov 2021 15:51:15.016 * Background append only file rewriting started by pid 11652
11648:S 28 Nov 2021 15:51:15.056 * AOF rewrite child asks to stop sending diffs.
11652:C 28 Nov 2021 15:51:15.056 * Parent agreed to stop sending diffs. Finalizing AOF...
11652:C 28 Nov 2021 15:51:15.056 * Concatenating 0.00 MB of AOF diff received from parent.
11652:C 28 Nov 2021 15:51:15.057 * SYNC append only file rewrite performed
11652:C 28 Nov 2021 15:51:15.057 * AOF rewrite: 2 MB of memory used by copy-on-write
11648:S 28 Nov 2021 15:51:15.087 * Background AOF rewrite terminated with success
11648:S 28 Nov 2021 15:51:15.087 * Residual parent diff successfully flushed to the rewritten AOF (0.00 MB)
11648:S 28 Nov 2021 15:51:15.087 * Background AOF rewrite finished successfully

  6. 命令测试

    1. 测试一：主节点写入数据，测试从节点是否可以获取(即主从复制)

    2. 测试二：从节点是否可以写入数据(即读写分离)

2 哨兵搭建

需求

  • 在上面主从复制架构的基础上，再在每台虚拟机上分别搭建一个哨兵，组成哨兵监控架构(如下图)。要求如下

    1. 模拟主节点宕机时的故障转移(使用 kill -9 pid 命令杀死主节点进行模拟)

搭建

  • 节点说明

    IP Redis节点
    192.168.192.130 6379 为当前 redis master 节点，26379 为 sentinel 节点
    192.168.192.131 6379 为当前 redis slave 节点，26379 为 sentinel 节点
    192.168.192.132 6379 为当前 redis slave 节点，26379 为 sentinel 节点

  1. 编写 sentinel.conf

    [root@localhost redis-5]# cd /usr/local/redis-5/conf/
[root@localhost conf]# vi sentinel-26379.conf
# 放行所有 IP 限制
bind 0.0.0.0
# 进程端口号
port 26379
# 后台启动
daemonize yes
# 日志记录文件
logfile "/usr/local/redis-5/log/sentinel-26379.log"
# 进程编号记录文件
pidfile /var/run/sentinel-26379.pid
# 指示 Sentinel 去监视一个名为 mymaster 的主服务器
sentinel monitor mymaster 192.168.192.130 6379 2
# 访问主节点的密码
sentinel auth-pass mymaster 123456
# Sentinel 认为服务器已经断线所需的毫秒数
sentinel down-after-milliseconds mymaster 10000
# 若 Sentinel 在该配置值内未能完成 failover 操作，则认为本次 failover 失败
sentinel failover-timeout mymaster 180000


# 哨兵sentinel的配置均一样,因此复制到另外两台机器即可
scp sentinel-26379.conf root@192.168.192.131:/usr/local/redis-5/conf/
scp sentinel-26379.conf root@192.168.192.132:/usr/local/redis-5/conf/

  2. 启动三个 Redis，若是已启动该步骤跳过

  3. 启动三个 Sentinel

    # 三台机器上分别执行如下命令
cd /usr/local/redis-5/
bin/redis-sentinel conf/sentinel-26379.conf

  4. 查看 Sentinel 是否启动成功

    # 使用 ps -ef |grep sentinel 命令是最简单的查看,仅能看是否启动成功,不能看sentinel各种信息

# 连接任意sentinel执行info sentinel 命令
[root@localhost redis-5]# bin/redis-cli -p 26379 info sentinel

sentinel_masters:1			# master节点个数
sentinel_tilt:0				# 是否进入TILT模式
sentinel_running_scripts:0		# 正在运行脚本个数
sentinel_scripts_queue_length:0		# 队列中正在排队的脚本个数
sentinel_simulate_failure_flags:0
# master节点name、state、address、slaves(从节点个数)，以及sentinels(哨兵个数)
master0:name=mymaster,status=ok,address=192.168.192.130:6379,slaves=2,sentinels=3

  5. 查看日志

    # 查看任意一台sentinel的日志即可,三者都差不多
[root@localhost log]# cd /usr/local/redis-5/log/
[root@localhost log]# tail -f -n 100 sentinel-26379.log 
11811:X 28 Nov 2021 16:28:19.463 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo
11811:X 28 Nov 2021 16:28:19.463 # Redis version=5.0.12, bits=64, commit=00000000, modified=0, pid=11811, just started
11811:X 28 Nov 2021 16:28:19.463 # Configuration loaded
11812:X 28 Nov 2021 16:28:19.466 * Increased maximum number of open files to 10032 (it was originally set to 1024).
11812:X 28 Nov 2021 16:28:19.466 * Running mode=sentinel, port=26379.
11812:X 28 Nov 2021 16:28:19.468 # Sentinel ID is afeb5e698bdf5c0c2bc7b1a6af5f1f7034a9e8a3
11812:X 28 Nov 2021 16:28:19.468 # +monitor master mymaster 192.168.192.130 6379 quorum 2
11812:X 28 Nov 2021 16:28:19.470 * +slave slave 192.168.192.131:6379 192.168.192.131 6379 @ mymaster 192.168.192.130 6379
11812:X 28 Nov 2021 16:28:19.471 * +slave slave 192.168.192.132:6379 192.168.192.132 6379 @ mymaster 192.168.192.130 6379
11812:X 28 Nov 2021 16:28:20.720 * +sentinel sentinel c4b37589df7a468b9d2b57584b7a0cf1f7b9ecfc 192.168.192.132 26379 @ mymaster 192.168.192.130 6379
11812:X 28 Nov 2021 16:28:21.094 * +sentinel sentinel 045878fd09c565628623d1f729236ca238de10fc 192.168.192.130 26379 @ mymaster 192.168.192.130 6379

  6. 模拟故障转移

    1. 故障转移前集群状态

      # 在主节点192.168.192.130上执行info replication命令
[root@localhost redis-5]# bin/redis-cli -a 123456 info replication

role:master				# 当前节点为主节点
connected_slaves:2		        # 两个从节点信息如下
slave0:ip=192.168.192.131,port=6379,state=online,offset=146720,lag=0
slave1:ip=192.168.192.132,port=6379,state=online,offset=146575,lag=1
master_replid:f06b2e1dcfb28e6648fc8d601413b61be295eca1
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:146720
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:146720

    2. 干掉主节点模拟故障转移

      [root@localhost redis-5]# ps -ef | grep redis
root       20090       1  0 15:48 ?        00:00:12 bin/redis-server 0.0.0.0:6379
root       20255       1  0 16:27 ?        00:00:08 bin/redis-sentinel 0.0.0.0:26379 [sentinel]
root       20290   19952  0 16:40 pts/0    00:00:00 grep redis
[root@localhost redis-5]# kill -9 20090

    3. 查看 sentinel 日志

      # 部分内容如下

# 1. 原主节点192.168.192.130 6379主观下线(sdown)
11812:X 28 Nov 2021 16:41:24.335 # +sdown master mymaster 192.168.192.130 6379
# 2. 选举新纪元开始,标号为1
11812:X 28 Nov 2021 16:41:24.407 # +new-epoch 1
# 3. 投票哪个sentinel执行故障转移
# 补充：当sentinel运行后,会自动生成一些内容(可使用命令 cat sentinel-26379.conf 查看),"045878fd09c565628623d1f729236ca238de10fc" 是"192.168.192.130 26379"机器上的sentinel
# 因此会由130机器上的sentinel进行故障转移,将转移后的结果(即现任主节点)通知给其他sentinel
11812:X 28 Nov 2021 16:41:24.408 # +vote-for-leader 045878fd09c565628623d1f729236ca238de10fc 1
# 4. 原主节点"192.168.192.130 6379"客观下线(quorum 2/2,即有一般以上的sentinel认为原主节点sdown了)
11812:X 28 Nov 2021 16:41:24.420 # +odown master mymaster 192.168.192.130 6379 #quorum 2/2
11812:X 28 Nov 2021 16:41:24.421 # Next failover delay: I will not start a failover before Sun Nov 28 16:47:25 2021
11812:X 28 Nov 2021 16:41:25.642 # +config-update-from sentinel 045878fd09c565628623d1f729236ca238de10fc 192.168.192.130 26379 @ mymaster 192.168.192.130 6379
# 5. 主节点由"192.168.192.130 6379"变为"192.168.192.131 6379"
11812:X 28 Nov 2021 16:41:25.643 # +switch-master mymaster 192.168.192.130 6379 192.168.192.131 6379
# 6. 通知 132 变为 131 的从节点(即告诉132,主节点现在变成131了)
11812:X 28 Nov 2021 16:41:25.644 * +slave slave 192.168.192.132:6379 192.168.192.132 6379 @ mymaster 192.168.192.131 6379
# 7. 通知 130 变成 131 的从节点(当130再次启动时,由原来的主节点现在变成131的从节点)
11812:X 28 Nov 2021 16:41:25.644 * +slave slave 192.168.192.130:6379 192.168.192.130 6379 @ mymaster 192.168.192.131 6379
# 8. 原主节点130下线,现在主节点变成131
11812:X 28 Nov 2021 16:41:35.677 # +sdown slave 192.168.192.130:6379 192.168.192.130 6379 @ mymaster 192.168.192.131 6379

    4. 原主节点130下线后再次查看集群状态

      # 在现主节点131上执行info replication命令
[root@localhost redis-5]# bin/redis-cli -a 123456 info replication

role:master			# 主节点
connected_slaves:1		# 一个从节点,信息如下
slave0:ip=192.168.192.132,port=6379,state=online,offset=421107,lag=1
master_replid:3a91a51191c0b94258caff79b04510318e6430db
master_replid2:f06b2e1dcfb28e6648fc8d601413b61be295eca1
master_repl_offset:421252
second_repl_offset:175178
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:421252

    5. 启动原主节点130，证明原主节点宕机后，再次启动会变成从节点

      # 0. 修改原主节点配置文件
# 从节点访问主节点密码(必须与 requirepass 一致)
masterauth 123456
# 从节点只读模式
replica-read-only yes

# 1. 启动
bin/redis-server conf/redis-6379.conf 

# 2. sentinel日志
# 重启从节点130,它现在的主节点时131
11812:X 28 Nov 2021 17:19:18.945 * +reboot slave 192.168.192.130:6379 192.168.192.130 6379 @ mymaster 192.168.192.131 6379
11812:X 28 Nov 2021 17:19:19.003 # -sdown slave 192.168.192.130:6379 192.168.192.130 6379 @ mymaster 192.168.192.131 6379

# 3. 查看集群状态
[root@localhost redis-5]# bin/redis-cli -a 123456 info replication

role:master
connected_slaves:2		# 从节点信息如下(原主节点130变为从节点)
slave0:ip=192.168.192.132,port=6379,state=online,offset=665891,lag=1
slave1:ip=192.168.192.130,port=6379,state=online,offset=665891,lag=1
master_replid:3a91a51191c0b94258caff79b04510318e6430db
master_replid2:f06b2e1dcfb28e6648fc8d601413b61be295eca1
master_repl_offset:666181
second_repl_offset:175178
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:666181

3 集群搭建

需求

  • 完成搭建三主三从的集群架构(如下图)。要求如下

    1. 六个 Redis 密码均为123456

搭建

  • 节点说明(下面只是搭建了6个Redis节点，主从待分配)

    IP Redis节点
    192.168.192.130 6371 端口，6372 端口
    192.168.192.131 6373 端口，6374 端口
    192.168.192.132 6375 端口，6376 端口

  1. 编写配置文件

    # (1) 6371 端口配置文件
[root@localhost conf]# cd /usr/local/redis-5/conf
[root@localhost conf]# vi redis-6371.conf 
bind 0.0.0.0
port 6371
daemonize yes
logfile "/usr/local/redis-5/log/redis-6371.log"
dir /usr/local/redis-5/data
dbfilename dump-6371.rdb
appendonly yes
appendfilename "appendonly-6371.aof"
requirepass 123456
# 从节点访问主节点密码(必须与 requirepass 一致)
masterauth 123456
# 是否开启集群模式，默认 no
cluster-enabled yes
# 集群节点信息文件，会保存在 dir 配置对应目录下
cluster-config-file nodes-6371.conf
# 集群节点连接超时时间
cluster-node-timeout 15000
# 集群节点 IP (可理解为当前机器的IP)
cluster-announce-ip 192.168.192.130
# 集群节点映射端口
cluster-announce-port 6371
# 集群节点总线端口 (基于当前端口 + 一万)
cluster-announce-bus-port 16371

# (2) 6372 端口配置文件
[root@localhost conf]# sed "s/6371/6372/g" redis-6371.conf > redis-6372.conf

# (3) 6373 端口配置文件(先把 6371.conf 配置文件传到 131 机器上，然后基于 6371.conf 进行修改)
[root@localhost conf]# scp redis-6371.conf root@192.168.192.131:/usr/local/redis-5/conf/
# 切换到131机器上进行如下操作
[root@localhost conf]# mv redis-6371.conf redis-6373.conf	
[root@localhost conf]# vi redis-6373.conf 	# 缩进部分为待修改内容
    # 集群节点 IP (可理解为当前机器的IP)
    cluster-announce-ip 192.168.192.131		# 修改为当期机器IP
    :%s/6371/6373/g				# 该命令在 vi 模式下进行全局替换
    
# (4) 6374 端口配置文件
[root@localhost conf]# sed "s/6373/6374/g" redis-6373.conf > redis-6374.conf

# (5) 6375 端口配置文件
[root@localhost conf]# scp redis-6373.conf root@192.168.192.132:/usr/local/redis-5/conf
[root@localhost conf]# mv redis-6373.conf redis-6375.conf
[root@localhost conf]# vi redis-6375.conf 
    cluster-announce-ip 192.168.192.132
    :%s/6373/6375/g					

# (6) 6376 端口配置文件
[root@localhost conf]# sed "s/6375/6376/g" redis-6375.conf > redis-6376.conf

  2. 启动六个redis节点

    [root@localhost redis-5]# cd /usr/local/redis-5/			# 当前为130机器
[root@localhost redis-5]# bin/redis-server conf/redis-6371.conf 
[root@localhost redis-5]# bin/redis-server conf/redis-6372.conf
[root@localhost redis-5]# bin/redis-server conf/redis-6373.conf		# 切换到131机器 
[root@localhost redis-5]# bin/redis-server conf/redis-6374.conf
[root@localhost redis-5]# bin/redis-server conf/redis-6375.conf		# 切换到132机器
[root@localhost redis-5]# bin/redis-server conf/redis-6376.conf

# 查看是否启动成功
[root@localhost redis-5]# ps -ef | grep redis		# cluster 表示集群模式,且启动成功
root        1766       1  0 17:55 ?        00:00:00 bin/redis-server 0.0.0.0:6375 [cluster]
root        1771       1  0 17:55 ?        00:00:00 bin/redis-server 0.0.0.0:6376 [cluster]
root        1779    1284  0 17:55 pts/0    00:00:00 grep redis

  3. 搭建集群

    • 记得关闭防火墙，若是基于上文的操作，到达此步骤时防火墙一般处于关闭状态，否则上面的主从、哨兵都过不了~
    [root@localhost redis-5]# bin/redis-cli -p 6371 -a 123456 --cluster create \
> 192.168.192.130:6371 192.168.192.130:6372 \
> 192.168.192.131:6373 192.168.192.131:6374 \
> 192.168.192.132:6375 192.168.192.132:6376 \
> --cluster-replicas 1

# 参数说明
# bin/redis-cli -p 6371 -a 123456 --cluster create ：连接任意一个客户端即可， --cluster create 表示创建集群
# "\" 感觉像连接符号，命令太长时使用。即 "--cluster create" 后面跟 6 个 redis 节点的ip 和 port
# --cluster-replicas 1 ，主节点数/从节点数的比例，使用1:1的比例，6 个节点最终会产生3主3从的集群环境

  4. 集群检查

    # (1) 检查集群状态(--cluster check命令),连接任意一个节点执行命令即可
[root@localhost redis-5]# bin/redis-cli -a 123456 --cluster check 192.168.192.130:6371
Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.
# 192.168.192.130:6371 是主节点,当前存了9个key,分配了5461个槽,有1个从节点(剩余两行信息同理)
192.168.192.130:6371 (ae7dcca6...) -> 0 keys | 5461 slots | 1 slaves.
192.168.192.131:6373 (22110c8e...) -> 0 keys | 5462 slots | 1 slaves.
192.168.192.132:6375 (dcfb83c5...) -> 0 keys | 5461 slots | 1 slaves.
[OK] 0 keys in 3 masters.
0.00 keys per slot on average.
# 集群状态检查操作由 192.168.192.130:6371 节点执行
>>> Performing Cluster Check (using node 192.168.192.130:6371)
# M 表示master即主节点,下面是它的信息和从节点个数
M: ae7dcca6168577aa5ce82ce3b1cd2c6736c99fa0 192.168.192.130:6371
   slots:[0-5460] (5461 slots) master
   1 additional replica(s)
# S 表示从节点,从节点信息以及复制的主节点ID.由此可知,该从节点的主节点为 6375 端口
S: 1161aafba5ba88d8917ec7e22a84dec242910eee 192.168.192.130:6372
   slots: (0 slots) slave
   replicates dcfb83c5ff905f0328a702c9a73a6bb08a2dabfa
M: 22110c8e2f5fbbce83dff59605c5afd0ccad7c2b 192.168.192.131:6373
   slots:[5461-10922] (5462 slots) master
   1 additional replica(s)
M: dcfb83c5ff905f0328a702c9a73a6bb08a2dabfa 192.168.192.132:6375
   slots:[10923-16383] (5461 slots) master
   1 additional replica(s)
S: 885400f76d5b7be0b10efb6767c0609a865f130e 192.168.192.131:6374
   slots: (0 slots) slave
   replicates ae7dcca6168577aa5ce82ce3b1cd2c6736c99fa0
S: 02fed433a6a5f0f45e6bf545f30036f9f41097fd 192.168.192.132:6376
   slots: (0 slots) slave
   replicates 22110c8e2f5fbbce83dff59605c5afd0ccad7c2b
[OK] All nodes agree about slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.

# (2) 查看集群信息(cluster info命令),连接任意一个节点执行命令即可
[root@localhost redis-5]# bin/redis-cli -h 192.168.192.131 -p 6373 -a 123456 cluster info
Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.
cluster_state:ok			# 集群状态
cluster_slots_assigned:16384	        # 集群槽已分配个数
cluster_slots_ok:16384			# 集群槽分配成功数
cluster_slots_pfail:0			# 集群槽可能失效数
cluster_slots_fail:0			# 集群槽已经失效数
cluster_known_nodes:6			# 集群中redis节点个数
cluster_size:3
cluster_current_epoch:6
cluster_my_epoch:3
cluster_stats_messages_ping_sent:922
cluster_stats_messages_pong_sent:956
cluster_stats_messages_meet_sent:1
cluster_stats_messages_sent:1879
cluster_stats_messages_ping_received:956
cluster_stats_messages_pong_received:923
cluster_stats_messages_received:1879

# (3) 查看节点信息(cluster nodes命令),连接任意一个节点执行命令即可
[root@localhost redis-5]# bin/redis-cli -h 192.168.192.131 -p 6373 -a 123456 cluster nodes
Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.
1161aafba5ba88d8917ec7e22a84dec242910eee 192.168.192.130:6372@16372 slave dcfb83c5ff905f0328a702c9a73a6bb08a2dabfa 0 1638094809000 5 connected
22110c8e2f5fbbce83dff59605c5afd0ccad7c2b 192.168.192.131:6373@16373 myself,master - 0 1638094807000 3 connected 5461-10922
885400f76d5b7be0b10efb6767c0609a865f130e 192.168.192.131:6374@16374 slave ae7dcca6168577aa5ce82ce3b1cd2c6736c99fa0 0 1638094807000 1 connected
02fed433a6a5f0f45e6bf545f30036f9f41097fd 192.168.192.132:6376@16376 slave 22110c8e2f5fbbce83dff59605c5afd0ccad7c2b 0 1638094809000 6 connected
ae7dcca6168577aa5ce82ce3b1cd2c6736c99fa0 192.168.192.130:6371@16371 master - 0 1638094810007 1 connected 0-5460
dcfb83c5ff905f0328a702c9a73a6bb08a2dabfa 192.168.192.132:6375@16375 master - 0 1638094808000 5 connected 10923-16383

  5. 查看主/从节点日志。略~

  6. 集群测试

    # 连接集群中的任意一个节点即可，但一定要加 "-c"，该参数表示集群模式。若是不加有可能会报 moved 异常~~
[root@localhost redis-5]# bin/redis-cli -c -a 123456 -h 192.168.192.130 -p 6371
Warning: Using a password with '-a' or '-u' option on the command line interface may not be safe.
192.168.192.130:6371> ping
PONG
192.168.192.130:6371> set zhangsan:age 18
-> Redirected to slot [9979] located at 192.168.192.131:6373
OK
192.168.192.131:6373> set zhangsan:phone 18362610001
-> Redirected to slot [3426] located at 192.168.192.130:6371
OK
192.168.192.130:6371> set zhangsan:address luoyang
-> Redirected to slot [6169] located at 192.168.192.131:6373
OK
192.168.192.131:6373> get zhangsan:phone
-> Redirected to slot [3426] located at 192.168.192.130:6371
"18362610001"
posted @ 2021-11-28 20:25  lkf-newlife  阅读(201)  评论(0)    收藏  举报