一、环境准备
| OS |
hostname |
IP |
role |
port |
| CentOS 7.6 |
redis01 |
10.4.7.100 |
master、slave |
6379、16379、6380、16380 |
| CentOS 7.6 |
redis02 |
10.4.7.101 |
master、slave |
6379、16379、6380、16380 |
| CentOS 7.6 |
redis03 |
10.4.7.102 |
master、slave |
6379、16379、6380、16380 |
说明：此集群采用三个主机，每个主机均运行两个redis实例，也就是说，三个主机上，共运行六个redis实例，构建为集群，集群采用三主三从模式！
集群默认使用redis监听端口+10000作为集群间通信端口！
二、安装redis
无特殊说明，在redis01机器上操作即可！
| # |
| $ yum install gcc* |
| $ wget http://download.redis.io/releases/redis-5.0.5.tar.gz |
| $ tar zxf redis-5.0.5.tar.gz;cd redis-5.0.5/ |
| $ make |
| # |
| $ mkdir /usr/local/redis/{bin,conf,data,logs} -p |
| $ mkdir /usr/local/redis/data/{6379,6380} |
| $ cp /root/redis-5.0.5/src/redis* /usr/local/redis/bin/ |
| $ cp /root/redis-5.0.5/redis.conf /usr/local/redis/conf/ |
| $ cd /usr/local/redis/bin/ |
| $ rm -f *.{h,c,o} |
三、修改配置文件
| # |
| $ cd /usr/local/redis/conf/ |
| # 可参考编译后生成的配置文件redis.conf，我这里就自己编写了 |
| $ vim redis_6379.conf |
| # 关于配置文件各项解释，可以自行搜索，或者看自带配置文件redis.conf中的注释 |
| port 6379 |
| bind 0.0.0.0 |
| daemonize yes |
| pidfile /var/run/redis_6379.pid |
| cluster-enabled yes |
| cluster-config-file nodes-6379.conf |
| cluster-node-timeout 25000 |
| |
| # rdb持久化相关配置 |
| save 900 1 |
| save 300 10 |
| save 60 10000 |
| dbfilename dump_6379.rdb |
| dir /usr/local/redis/data/6379 |
| stop-writes-on-bgsave-error no |
| rdbcompression yes |
| rdbchecksum yes |
| rdb-save-incremental-fsync yes |
| |
| |
| # aof持久化相关配置 |
| appendonly yes |
| appendfilename "appendonly_6379.aof" |
| appendfsync everysec |
| no-appendfsync-on-rewrite no |
| auto-aof-rewrite-percentage 100 |
| auto-aof-rewrite-min-size 64mb |
| aof-rewrite-incremental-fsync yes |
| |
| |
| loglevel notice |
| logfile /usr/local/redis/logs/redis-6379.logs |
| maxclients 15000 |
| maxmemory 20gb |
| maxmemory-policy volatile-lru |
| protected-mode no |
| |
| # 主从复制相关配置 |
| replica-serve-stale-data yes |
| replica-read-only yes |
| repl-diskless-sync no |
| repl-diskless-sync-delay 5 |
| repl-disable-tcp-nodelay no |
| replica-priority 100 |
| repl-backlog-size 10mb |
| repl-timeout 120 |
| |
| # 集群密码认证（集群中所有节点密码必须一致，可不开启） |
| masterauth "123456" |
| requirepass "123456" |
| |
| # |
| $ cp redis_6379.conf redis_6380.conf |
| $ sed -i 's/6379/6380/g' redis_6380.conf |
将redis相关文件分发到其余两个机器上
| $ for i in 10.4.7.10{1..2};do scp -r /usr/local/redis ${i}:/usr/local/;done |
设置命令路径的环境变量（集群中所有集群都做）
| $ echo 'export PATH=$PATH:/usr/local/redis/bin/' >> /etc/profile |
| $ source /etc/profile |
| $ redis-server -v |
| Redis server v=5.0.5 sha=00000000:0 malloc=jemalloc-5.1.0 bits=64 build=526c61b9b59aa5bc |
至此，三个主机共6个实例的配置文件就准备好了，可以启动了（但是启动之前需要修复一下启动过程中日志产生的warning事件）。
四、解决启动过程中日志产生的warning事件
以下操作需要在三台主机上都执行一下！
4.1 解决最大打开文件数问题
| $ ulimit -n |
| 1024 |
| $ echo '* - nofile 65535' >> /etc/security/limits.conf |
| # 修改后，重新登录即可生效，重新登录后再次查看当前值 |
| $ ulimit -n |
| 65535 |
4.2 解决TCP积压值过小问题
| $ echo "net.core.somaxconn = 65535" > /etc/sysctl.d/redis.conf |
| $ sysctl -p /etc/sysctl.d/redis.conf |
| net.core.somaxconn = 1024 |
4.3 允许分配所有的物理内存
| $ echo "vm.overcommit_memory = 1" >> /etc/sysctl.d/redis.conf |
| $ sysctl -p /etc/sysctl.d/redis.conf |
| net.core.somaxconn = 1024 |
| vm.overcommit_memory = 1 |
4.4 解决内存透明大页警告warning问题
| $ echo never > /sys/kernel/mm/transparent_hugepage/enabled |
| #上述指令只是当前生效，重启后就会失效，接下来改为永久生效 |
| $ echo 'echo never > /sys/kernel/mm/transparent_hugepage/enabled' >> /etc/rc.local |
| $ chmod +x /etc/rc.d/rc.local |
至此，reboot重启服务器也好，只是重启redis服务也好，都不会再报哪些warning问题了！
五、启动redis
| $ redis-server /usr/local/redis/conf/redis_6379.conf |
| $ redis-server /usr/local/redis/conf/redis_6380.conf |
| $ ss -antp | grep 63 |
| LISTEN 0 511 *:6379 *:* users:(("redis-server",pid=13273,fd=6)) |
| LISTEN 0 511 *:6380 *:* users:(("redis-server",pid=13292,fd=6)) |
| LISTEN 0 511 *:16379 *:* users:(("redis-server",pid=13273,fd=9)) |
| LISTEN 0 511 *:16380 *:* users:(("redis-server",pid=13292,fd=9)) |
| $ pkill -9 redis-server |
六、创建集群
6.1 取消集群密码
如果你的集群有密码，则需进行此步骤，若没有设置密码，则可忽略。此脚本在后面恢复集群密码，以及对redis集群扩缩容，都会用到。
| $ cat redis_setpass.sh |
| #!/usr/bin/env bash |
| # 定义参与集群的IP |
| IPS=( |
| 10.4.7.100 |
| 10.4.7.101 |
| 10.4.7.102 |
| ) |
| |
| # 定义集群密码 |
| PASS='123456' |
| |
| # 定义每个节点的监听端口 |
| PORTS=( |
| 6379 |
| 6380 |
| ) |
| |
| |
| # delete password |
| del_pass() { |
| for ip in ${IPS[@]} |
| do |
| for port in ${PORTS[@]} |
| do |
| redis-cli -c -h $ip -p $port -a $PASS config set masterauth "" |
| redis-cli -c -h $ip -p $port -a $PASS config set requirepass "" |
| done |
| echo "$ip delete password" |
| done |
| } |
| |
| |
| add_pass() { |
| for ip in ${IPS[@]} |
| do |
| for port in ${PORTS[@]} |
| do |
| redis-cli -c -h $ip -p $port config set masterauth "$PASS" |
| redis-cli -c -h $ip -p $port config set requirepass "$PASS" |
| done |
| echo "$ip add password" |
| done |
| } |
| |
| |
| env=$1 |
| |
| if [[ ${env} == "del" ]];then |
| echo "del redis password" |
| del_pass |
| |
| elif [[ ${env} == "add" ]];then |
| add_pass |
| |
| else |
| echo "${env} not add || del " |
| echo ' exit ..' |
| exit |
| fi |
| |
| $ chmod +x redis_setpass.sh |
| $ ./redis_setpass.sh del |
6.2 创建集群
以下操作在任意一台主机上进行即可！
| $ redis-cli --cluster create 10.4.7.100:6379 10.4.7.100:6380 10.4.7.101:6379 10.4.7.101:6380 10.4.7.102:6379 10.4.7.102:6380 --cluster-replicas 1 |
| # 创建集群，需指定参与集群的实例，并指定replicas副本数为1 |
6.3 测试集群
| $ redis-cli -c -h 10.4.7.100 -p 6379 |
| |
| 10.4.7.100:6379> cluster nodes # 查看master及slave之间的对照关系 |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612245916000 5 connected 10923-16383 |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 myself,master - 0 1612245915000 1 connected 0-5460 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612245916000 6 connected |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612245915000 4 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612245916740 5 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612245913000 3 connected 5461-10922 |
| |
| 10.4.7.100:6379> cluster info # 查看集群状态 |
| 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:6 |
| cluster_my_epoch:1 |
| cluster_stats_messages_ping_sent:6647 |
| cluster_stats_messages_pong_sent:6778 |
| cluster_stats_messages_sent:13425 |
| cluster_stats_messages_ping_received:6773 |
| cluster_stats_messages_pong_received:6647 |
| cluster_stats_messages_meet_received:5 |
| cluster_stats_messages_received:13425 |
| |
| # 查询key值 |
| 10.4.7.100:6379> set name zz # 设置key |
| -> Redirected to slot [5798] located at 10.4.7.101:6379 |
| OK |
| # 可以发现下面的命令提示符都自动换成了10.4.7.101:6379这个实例的 |
| 10.4.7.101:6379> get name |
| "zz" |
| |
| # 登录到其他节点，查看是否可以查询到刚刚的key值 |
| $ redis-cli -c -p 6380 -h 10.4.7.102 |
| 10.4.7.102:6380> get name |
| -> Redirected to slot [5798] located at 10.4.7.101:6379 |
| "zz" |
至于其高可用，及master和salve的故障自动切换，自行测试即可。正常情况下，master宕机后，slave会自动顶上成为新的master。待master恢复后，会变成新master节点的slave。
6.4 使用脚本恢复集群密码
6.5 重建redis集群
若需要重建redis集群，则需要删除rdb、aof这两种数据持久化文件（若没开启aof，则不会有aof持久化文件），以及群集配置文件，在此博文中，需要删除的文件如下：
| $ tree /usr/local/redis/data/ |
| /usr/local/redis/data/ |
| ├── 6379 |
| │ ├── appendonly_6379.aof |
| │ ├── dump_6379.rdb |
| │ └── nodes-6379.conf |
| └── 6380 |
| ├── appendonly_6380.aof |
| ├── dump_6380.rdb |
| └── nodes-6380.conf |
| |
| 2 directories, 6 files |
| # 将数据文件删除后，重新执行redis 创建集群的命令即可！ |
| $ redis-cli --cluster create 10.4.7.100:6379 10.4.7.100:6380 10.4.7.101:6379 10.4.7.101:6380 10.4.7.102:6379 10.4.7.102:6380 --cluster-replicas 1 |
七、集群扩容
当现有redis集群无法满足业务需求，需要扩容的话，可以按照如下进行配置（redis较高版本和低版本的扩容方式不太一样，如果你的集群是 3.x.x的，或者redis-cli 命令不支持集群扩容操）。
7.1 部署新的redis节点
先准备好要加入集群中的redis实例，我这里的IP为 10.4.7.103，上面启动了两个实例，分别是 6379 和 6380。
| $ scp -r /usr/local/redis 10.4.7.103:/usr/local/ |
| |
| $ rm -rf /usr/local/redis/data/*/* |
| $ rm -rf /usr/local/redis/logs/* |
| |
| # |
| $ echo '* - nofile 65535' >> /etc/security/limits.conf |
| # 修改后，重新登录即可生效，重新登录后再次查看当前值 |
| $ ulimit -n |
| 65535 |
| |
| $ cat >> /etc/sysctl.d/redis.conf << EOF |
| net.core.somaxconn = 65535 |
| vm.overcommit_memory = 1 |
| EOF |
| |
| $ sysctl -p /etc/sysctl.d/redis.conf #刷新使其生效 |
| |
| |
| $ echo never > /sys/kernel/mm/transparent_hugepage/enabled |
| |
| $ echo 'echo never > /sys/kernel/mm/transparent_hugepage/enabled' >> /etc/rc.local |
| $ chmod +x /etc/rc.d/rc.local |
| |
| # 配置环境变量 |
| $ echo 'export PATH=$PATH:/usr/local/redis/bin/' >> /etc/profile |
| $ source /etc/profile |
| $ redis-server -v # 查看redis版本 |
| Redis server v=5.0.5 sha=00000000:0 malloc=jemalloc-5.1.0 bits=64 build=37a632ba3989f893 |
7.2 启动新节点的redis实例
| $ redis-server /usr/local/redis/conf/redis_6379.conf |
| $ redis-server /usr/local/redis/conf/redis_6380.conf |
| $ ss -anpt | grep 63 |
| LISTEN 0 511 *:6379 *:* users:(("redis-server",pid=8302,fd=6)) |
| LISTEN 0 511 *:6380 *:* users:(("redis-server",pid=8321,fd=6)) |
| LISTEN 0 511 *:16379 *:* users:(("redis-server",pid=8302,fd=9)) |
| LISTEN 0 511 *:16380 *:* users:(("redis-server",pid=8321,fd=9)) |
7.3 取消redis集群密码
| # 脚本中需增加刚刚新增的redis的IP及端口（自行更改） |
| $ ./redis_setpass.sh del |
7.4 redis新节点加入集群
| # 以下操作在任意一台可以连接到集群的节点上进行即可（10.4.7.100:6379 为集群中已存在的节点） |
| $ redis-cli --cluster add-node 10.4.7.103:6379 10.4.7.100:6379 |
| $ redis-cli --cluster add-node 10.4.7.103:6380 10.4.7.100:6379 |
7.5 查询现有节点信息
| $ redis-cli -c -h 10.4.7.100 -p 6379 cluster nodes |
可以看到，10.4.7.103 的两个实例都已加入(都为master)，但还没有分配slots。
7.6 给新加入的master分配solts
| $ redis-cli --cluster reshard 10.4.7.100:6379 |
执行上述命令后，操作如下：
如果在下面选择 all 的话，请注意，你如果是计划扩容多个master到集群中，那么你每次需要指定的要分配的slots是不一样的。
假设集群现有master为 3个，那么增加第四个master时，要分配的slots为：4096，如果继续增加第五个master，那么此时要分配的slots就应该为：16384 / 5 = 3276
7.7 给新加入的master分配slave
| $ redis-cli -h 10.4.7.103 -p 6380 -c |
| 10.4.7.103:6380> cluster nodes # 查询节点信息（找到那个没有从节点的 master的ID） |
| 8a0605ab116aa6ce911468feeadafbe68440fb46 10.4.7.103:6380@16380 myself,master - 0 1612248723000 0 connected |
| 210174dd1a76e9d77b3348854dcfb97223374496 10.4.7.103:6379@16379 master - 0 1612248724000 7 connected 0-1364 5461-6826 10923-12287 |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612248724004 5 connected 12288-16383 |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 master - 0 1612248726019 1 connected 1365-5460 |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612248725012 1 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612248724000 5 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612248722997 3 connected 6827-10922 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612248724000 3 connected |
| |
| # 注，由于我只增加了一台机器上的redis，主从都在一个机器上，所以只能这样 |
| # 如果你要扩容多个节点，那么最好将slave和master分别放在不同机器上 |
| |
| 10.4.7.103:6380> CLUSTER REPLICATE 210174dd1a76e9d77b3348854dcfb97223374496 |
| # 指定复制 新master |
| |
| 10.4.7.103:6380> cluster nodes # 确认集群中的master和slave无异常 |
| 8a0605ab116aa6ce911468feeadafbe68440fb46 10.4.7.103:6380@16380 myself,slave 210174dd1a76e9d77b3348854dcfb97223374496 0 1612248895000 0 connected |
| 210174dd1a76e9d77b3348854dcfb97223374496 10.4.7.103:6379@16379 master - 0 1612248895000 7 connected 0-1364 5461-6826 10923-12287 |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612248898000 5 connected 12288-16383 |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 master - 0 1612248898347 1 connected 1365-5460 |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612248895000 1 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612248897000 5 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612248897338 3 connected 6827-10922 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612248895326 3 connected |
7.8 集群恢复密码
八、集群缩容
假设我们要将上面新增的 10.4.7.103 的6379和6380两个实例从集群中移除，那么要怎样操作呢？
便于操作，先把集群密码取消！
8.1 移除slave
| $ redis-cli -c -h 10.4.7.100 -p 6379 cluster nodes |
| # 我这次需要移除的slaveID为：8a0605ab116aa6ce911468feeadafbe68440fb46 |
| 210174dd1a76e9d77b3348854dcfb97223374496 10.4.7.103:6379@16379 master - 0 1612249324000 7 connected 0-1364 5461-6826 10923-12287 |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612249324000 5 connected 12288-16383 |
| 8a0605ab116aa6ce911468feeadafbe68440fb46 10.4.7.103:6380@16380 slave 210174dd1a76e9d77b3348854dcfb97223374496 0 1612249327362 7 connected |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 myself,master - 0 1612249324000 1 connected 1365-5460 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612249325346 6 connected |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612249326351 4 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612249324339 5 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612249325000 3 connected 6827-10922 |
| |
| # 开始移除 |
| $ redis-cli --cluster del-node 10.4.7.100:6379 8a0605ab116aa6ce911468feeadafbe68440fb46 |
| >>> Removing node 8a0605ab116aa6ce911468feeadafbe68440fb46 from cluster 10.4.7.100:6379 |
| >>> Sending CLUSTER FORGET messages to the cluster... |
| >>> SHUTDOWN the node. |
| # 从输出信息可以看到已经移除成功了，并且要下线的redis实例已被停止。 |
| |
| $ redis-cli -c -h 10.4.7.100 -p 6379 cluster nodes |
| 210174dd1a76e9d77b3348854dcfb97223374496 10.4.7.103:6379@16379 master - 0 1612249527000 7 connected 0-1364 5461-6826 10923-12287 |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612249529000 5 connected 12288-16383 |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 myself,master - 0 1612249530000 1 connected 1365-5460 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612249527000 6 connected |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612249529931 4 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612249528924 5 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612249530940 3 connected 6827-10922 |
8.2 移除master
我们尝试删除之前加入的主节点6379，这个步骤相对比较麻烦一些，因为主节点的里面是有分配了hash槽的，所以我们这里必须先把6379里的hash槽放入到其他的可用主节点中去，然后再进行移除节点操作，不然会出现数据丢失问题（最好将要下线的master的slots数量平均分配至其他master上，所以只能一次分配部分solts槽，有几个master就要分配几次）：
| $ redis-cli --cluster reshard 10.4.7.100:6379 |
按照上述方式，多分配几次，直到将要下线的master的solts完全分配出去。
| $ redis-cli -c -h 10.4.7.100 -p 6379 cluster nodes |
| 210174dd1a76e9d77b3348854dcfb97223374496 10.4.7.103:6379@16379 master - 0 1612250066245 7 connected |
| c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379@16379 master - 0 1612250064000 9 connected 5461-6825 12288-16383 |
| 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379@16379 myself,master - 0 1612250065000 8 connected 0-5460 |
| 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380@16380 slave 22311cc5373dff550d373f30b05dd41f2be50838 0 1612250068261 10 connected |
| a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380@16380 slave 3e10873c57baf06acad788b815344edecc6b3428 0 1612250066000 8 connected |
| 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380@16380 slave c81b92bf2a1ea24c0881c296176f6a48cdce4460 0 1612250067253 9 connected |
| 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379@16379 master - 0 1612250065238 10 connected 6826-12287 |
| |
| # 移除要下线的主机点 |
| $ redis-cli --cluster del-node 10.4.7.100:6379 210174dd1a76e9d77b3348854dcfb97223374496 |
| >>> Removing node 210174dd1a76e9d77b3348854dcfb97223374496 from cluster 10.4.7.100:6379 |
| >>> Sending CLUSTER FORGET messages to the cluster... |
| >>> SHUTDOWN the node. |
至此，那台10.4.7.103 那台机器可以光荣退休了！
8.3 检查集群状态
| $ redis-cli --cluster check 10.4.7.100:6379 |
| 10.4.7.100:6379 (3e10873c...) -> 0 keys | 5461 slots | 1 slaves. |
| 10.4.7.102:6379 (c81b92bf...) -> 1 keys | 5461 slots | 1 slaves. |
| 10.4.7.101:6379 (22311cc5...) -> 0 keys | 5462 slots | 1 slaves. |
| [OK] 1 keys in 3 masters. |
| 0.00 keys per slot on average. |
| >>> Performing Cluster Check (using node 10.4.7.100:6379) |
| M: 3e10873c57baf06acad788b815344edecc6b3428 10.4.7.100:6379 |
| slots:[0-5460] (5461 slots) master |
| 1 additional replica(s) |
| M: c81b92bf2a1ea24c0881c296176f6a48cdce4460 10.4.7.102:6379 |
| slots:[5461-6825],[12288-16383] (5461 slots) master |
| 1 additional replica(s) |
| S: 82ff3b96e4f9bbe5e723fd066fe593294602bb02 10.4.7.102:6380 |
| slots: (0 slots) slave |
| replicates 22311cc5373dff550d373f30b05dd41f2be50838 |
| S: a48d503ce893f4e4b3f58a81fd2380dde358db48 10.4.7.101:6380 |
| slots: (0 slots) slave |
| replicates 3e10873c57baf06acad788b815344edecc6b3428 |
| S: 95f8bd253511c7c177e5d5f4caa5367ab85ca54a 10.4.7.100:6380 |
| slots: (0 slots) slave |
| replicates c81b92bf2a1ea24c0881c296176f6a48cdce4460 |
| M: 22311cc5373dff550d373f30b05dd41f2be50838 10.4.7.101:6379 |
| slots:[6826-12287] (5462 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. |
8.4 集群恢复密码
| # 当然需要把刚才移除的节点的IP地址从脚本删除 |
| $ ./redis_setpass.sh add |
原文链接
【推荐】国内首个AI IDE，深度理解中文开发场景，立即下载体验Trae
【推荐】编程新体验，更懂你的AI，立即体验豆包MarsCode编程助手
【推荐】抖音旗下AI助手豆包，你的智能百科全书，全免费不限次数
【推荐】轻量又高性能的 SSH 工具 IShell：AI 加持，快人一步
· AI与.NET技术实操系列：向量存储与相似性搜索在 .NET 中的实现
· 基于Microsoft.Extensions.AI核心库实现RAG应用
· Linux系列：如何用heaptrack跟踪.NET程序的非托管内存泄露
· 开发者必知的日志记录最佳实践
· SQL Server 2025 AI相关能力初探
· 震惊！C++程序真的从main开始吗？99%的程序员都答错了
· 【硬核科普】Trae如何「偷看」你的代码？零基础破解AI编程运行原理
· 单元测试从入门到精通
· 上周热点回顾（3.3-3.9）
· winform 绘制太阳，地球，月球 运作规律