Redis的Cluster配置
作者：尹正杰
版权声明：原创作品，谢绝转载！否则将追究法律责任。
一.安装Redis并启动
1>.修改Redis的配置文件（本实验只有三个节点）
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
bind node101.yinzhengjie.org.cn
protected-mode yes
port
tcp-backlog
unixsocket "/tmp/redis.sock"
timeout
tcp-keepalive
daemonize no
supervised systemd
pidfile "/var/run/redis_6379.pid"
loglevel notice
logfile "/var/log/redis/redis.log"
databases
save
save
save
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename "dump.rdb"
dir "/var/lib/redis"
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay
repl-disable-tcp-nodelay no
slave-priority
requirepass "yinzhengjie"
appendonly no
appendfilename "appendonly.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit
cluster-enabled yes #我们需要启用cluster功能
cluster-config-file redis-cluster.conf #指定cluster的配置文件
cluster-node-timeout #指定节点的超时时间
slowlog-log-slower-than
slowlog-max-len
latency-monitor-threshold
notify-keyspace-events ""
hash-max-ziplist-entries
hash-max-ziplist-value
list-max-ziplist-size -
list-compress-depth
set-max-intset-entries
zset-max-ziplist-entries
zset-max-ziplist-value
hll-sparse-max-bytes
activerehashing yes
client-output-buffer-limit normal
client-output-buffer-limit slave 256mb 64mb
client-output-buffer-limit pubsub 32mb 8mb
hz
aof-rewrite-incremental-fsync yes
[[email protected] ~]#
[[email protected] ~]#
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
[[email protected] ~]#
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
bind node102.yinzhengjie.org.cn
protected-mode yes
port
tcp-backlog
unixsocket "/tmp/redis.sock"
timeout
tcp-keepalive
daemonize no
supervised systemd
pidfile "/var/run/redis_6379.pid"
loglevel notice
logfile "/var/log/redis/redis.log"
databases
save
save
save
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename "dump.rdb"
dir "/var/lib/redis"
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay
repl-disable-tcp-nodelay no
slave-priority
requirepass "yinzhengjie"
appendonly no
appendfilename "appendonly.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit
cluster-enabled yes
cluster-config-file redis-cluster.conf
cluster-node-timeout
slowlog-log-slower-than
slowlog-max-len
latency-monitor-threshold
notify-keyspace-events ""
hash-max-ziplist-entries
hash-max-ziplist-value
list-max-ziplist-size -
list-compress-depth
set-max-intset-entries
zset-max-ziplist-entries
zset-max-ziplist-value
hll-sparse-max-bytes
activerehashing yes
client-output-buffer-limit normal
client-output-buffer-limit slave 256mb 64mb
client-output-buffer-limit pubsub 32mb 8mb
hz
aof-rewrite-incremental-fsync yes
[[email protected] ~]#
[[email protected] ~]#
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
bind node103.yinzhengjie.org.cn
protected-mode yes
port
tcp-backlog
unixsocket "/tmp/redis.sock"
timeout
tcp-keepalive
daemonize no
supervised systemd
pidfile "/var/run/redis_6379.pid"
loglevel notice
logfile "/var/log/redis/redis.log"
databases
save
save
save
stop-writes-on-bgsave-error yes
rdbcompression yes
rdbchecksum yes
dbfilename "dump.rdb"
dir "/var/lib/redis"
slave-serve-stale-data yes
slave-read-only yes
repl-diskless-sync no
repl-diskless-sync-delay
repl-disable-tcp-nodelay no
slave-priority
requirepass "yinzhengjie"
appendonly no
appendfilename "appendonly.aof"
appendfsync everysec
no-appendfsync-on-rewrite no
auto-aof-rewrite-percentage
auto-aof-rewrite-min-size 64mb
aof-load-truncated yes
lua-time-limit
cluster-enabled yes
cluster-config-file redis-cluster.conf
cluster-node-timeout
slowlog-log-slower-than
slowlog-max-len
latency-monitor-threshold
notify-keyspace-events ""
hash-max-ziplist-entries
hash-max-ziplist-value
list-max-ziplist-size -
list-compress-depth
set-max-intset-entries
zset-max-ziplist-entries
zset-max-ziplist-value
hll-sparse-max-bytes
activerehashing yes
client-output-buffer-limit normal
client-output-buffer-limit slave 256mb 64mb
client-output-buffer-limit pubsub 32mb 8mb
hz
aof-rewrite-incremental-fsync yes
[[email protected] ~]#
[[email protected] ~]#
[[email protected] ~]# egrep -v "^#|^$" /etc/redis.conf
2>.启动Redis集群
[[email protected] ~]# systemctl start redis
[[email protected] ~]#
[[email protected] ~]# systemctl status redis
● redis.service - Redis persistent key-value database
Loaded: loaded (/usr/lib/systemd/system/redis.service; disabled; vendor preset: disabled)
Drop-In: /etc/systemd/system/redis.service.d
└─limit.conf
Active: active (running) since Sun -- :: CST; 5min ago
Main PID: (redis-server)
CGroup: /system.slice/redis.service
└─ /usr/bin/redis-server node101.yinzhengjie.org.cn: [cluster] Apr :: node101.yinzhengjie.org.cn systemd[]: Starting Redis persistent key-value database...
Apr :: node101.yinzhengjie.org.cn systemd[]: Started Redis persistent key-value database.
[[email protected] ~]#
[[email protected] ~]#
[[email protected] ~]# systemctl start redis
[[email protected] ~]# systemctl start redis
[[email protected] ~]#
[[email protected] ~]# systemctl status redis
● redis.service - Redis persistent key-value database
Loaded: loaded (/usr/lib/systemd/system/redis.service; disabled; vendor preset: disabled)
Drop-In: /etc/systemd/system/redis.service.d
└─limit.conf
Active: active (running) since Sun -- :: CST; 3s ago
Main PID: (redis-server)
CGroup: /system.slice/redis.service
└─ /usr/bin/redis-server node102.yinzhengjie.org.cn: [cluster] Apr :: node102.yinzhengjie.org.cn systemd[]: Starting Redis persistent key-value database...
Apr :: node102.yinzhengjie.org.cn systemd[]: Started Redis persistent key-value database.
[[email protected] ~]#
[[email protected] ~]# systemctl start redis
[[email protected] ~]# systemctl start redis
[[email protected] ~]#
[[email protected] ~]# systemctl status redis
● redis.service - Redis persistent key-value database
Loaded: loaded (/usr/lib/systemd/system/redis.service; disabled; vendor preset: disabled)
Drop-In: /etc/systemd/system/redis.service.d
└─limit.conf
Active: active (running) since Sat -- :: PDT; 3s ago
Main PID: (redis-server)
CGroup: /system.slice/redis.service
└─ /usr/bin/redis-server node103.yinzhengjie.org.cn: [cluster] Apr :: node103.yinzhengjie.org.cn systemd[]: Starting Redis persistent key-value database...
Apr :: node103.yinzhengjie.org.cn systemd[]: Started Redis persistent key-value database.
[[email protected] ~]#
[[email protected] ~]#
[[email protected] ~]# systemctl start redis
3>.登录Redis服务器测试
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:>
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie
二.配置Redis Cluster
1>.查看cluster相关命令的帮助信息
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:> HELP @cluster CLUSTER ADDSLOTS slot [slot ...]
summary: Assign new hash slots to receiving node
since: 3.0. CLUSTER COUNT-FAILURE-REPORTS node-id
summary: Return the number of failure reports active for a given node
since: 3.0. CLUSTER COUNTKEYSINSLOT slot
summary: Return the number of local keys in the specified hash slot
since: 3.0. CLUSTER DELSLOTS slot [slot ...]
summary: Set hash slots as unbound in receiving node
since: 3.0. CLUSTER FAILOVER [FORCE|TAKEOVER]
summary: Forces a slave to perform a manual failover of its master.
since: 3.0. CLUSTER FORGET node-id
summary: Remove a node from the nodes table
since: 3.0. CLUSTER GETKEYSINSLOT slot count
summary: Return local key names in the specified hash slot
since: 3.0. CLUSTER INFO -
summary: Provides info about Redis Cluster node state
since: 3.0. CLUSTER KEYSLOT key
summary: Returns the hash slot of the specified key
since: 3.0. CLUSTER MEET ip port
summary: Force a node cluster to handshake with another node
since: 3.0. CLUSTER NODES -
summary: Get Cluster config for the node
since: 3.0. CLUSTER REPLICATE node-id
summary: Reconfigure a node as a slave of the specified master node
since: 3.0. CLUSTER RESET [HARD|SOFT]
summary: Reset a Redis Cluster node
since: 3.0. CLUSTER SAVECONFIG -
summary: Forces the node to save cluster state on disk
since: 3.0. CLUSTER SET-CONFIG-EPOCH config-epoch
summary: Set the configuration epoch in a new node
since: 3.0. CLUSTER SETSLOT slot IMPORTING|MIGRATING|STABLE|NODE [node-id]
summary: Bind a hash slot to a specific node
since: 3.0. CLUSTER SLAVES node-id
summary: List slave nodes of the specified master node
since: 3.0. CLUSTER SLOTS -
summary: Get array of Cluster slot to node mappings
since: 3.0. READONLY -
summary: Enables read queries for a connection to a cluster slave node
since: 3.0. READWRITE -
summary: Disables read queries for a connection to a cluster slave node
since: 3.0. node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:6379> HELP @cluster
2>.命令行分配切片范围
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie -c cluster addslots {..}
OK
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node102.yinzhengjie.org.cn -a yinzhengjie -c cluster addslots {..}
OK
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node103.yinzhengjie.org.cn -a yinzhengjie -c cluster addslots {..}
OK
[[email protected] ~]#
[[email protected] ~]#
3>.互相暴露各个节点
[[email protected] ~]# redis-cli -h node103.yinzhengjie.org.cn -a yinzhengjie
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> CLUSTER INFO
cluster_state:fail
cluster_slots_assigned:
cluster_slots_ok:
cluster_slots_pfail:
cluster_slots_fail:
cluster_known_nodes:
cluster_size:
cluster_current_epoch:
cluster_my_epoch:
cluster_stats_messages_sent:
cluster_stats_messages_received:
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:6379> CLUSTER INFO
node103.yinzhengjie.org.cn:> CLUSTER MEET 172.30.1.101
OK
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> CLUSTER MEET 172.30.1.102
OK
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> CLUSTER INFO
cluster_state:ok
cluster_slots_assigned:
cluster_slots_ok:
cluster_slots_pfail:
cluster_slots_fail:
cluster_known_nodes:
cluster_size:
cluster_current_epoch:
cluster_my_epoch:
cluster_stats_messages_sent:
cluster_stats_messages_received:
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:6379> CLUSTER INFO
[[email protected] ~]# redis-cli -h node102.yinzhengjie.org.cn -a yinzhengjie
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:> CLUSTER INFO
cluster_state:ok
cluster_slots_assigned:
cluster_slots_ok:
cluster_slots_pfail:
cluster_slots_fail:
cluster_known_nodes:
cluster_size:
cluster_current_epoch:
cluster_my_epoch:
cluster_stats_messages_sent:
cluster_stats_messages_received:
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:6379> CLUSTER INFO
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:> CLUSTER INFO
cluster_state:ok
cluster_slots_assigned:
cluster_slots_ok:
cluster_slots_pfail:
cluster_slots_fail:
cluster_known_nodes:
cluster_size:
cluster_current_epoch:
cluster_my_epoch:
cluster_stats_messages_sent:
cluster_stats_messages_received:
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:6379> CLUSTER INFO
4>.验证Redis的写操作
[[email protected] ~]# redis-cli -h node101.yinzhengjie.org.cn -a yinzhengjie
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:> SET name Jason #注意，我们在node101.yinzhengjie.org.cn创建数据时失败啦！返回的是错误，说是name这个字段被hash到5798这个分片上，而该分片在172.30.1.102:6379这个节点上，因此我们应该去该节点配置！
(error) MOVED 172.30.1.102:
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:>
node101.yinzhengjie.org.cn:> quit
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node102.yinzhengjie.org.cn -a yinzhengjie #于是，按照其错误的提示，我们登录到172.30.1.102:6379这个节点上去设置KEY
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:> SET name Jason #果不其然，在该节点就可以成功创建名称为name的KEY
OK
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:> KEYS *
) "name"
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:>
5>.验证Redis的读操作
[[email protected] ~]# redis-cli -h node103.yinzhengjie.org.cn -a yinzhengjie
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> KEYS *
(empty list or set)
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> GET name #当我们想要读取name字段对应的value时，发现当前节点可以帮我们路由到真正存储的该节点上去
(error) MOVED 172.30.1.102:
node103.yinzhengjie.org.cn:>
node103.yinzhengjie.org.cn:> quit
[[email protected] ~]#
[[email protected] ~]# redis-cli -h node102.yinzhengjie.org.cn -a yinzhengjie
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:> KEYS *
) "name"
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:> GET name #按照上面的报错提示登录到指定的节点后，发现是可以读取数据的！
"Jason"
node102.yinzhengjie.org.cn:>
node102.yinzhengjie.org.cn:>
经测试，读写操作虽然能实现，但是对于程序员来说，手动去解决这些问题简直太麻烦了，推荐大家使用Cerberus（芒果TV）的解决方案，可以帮我们轻松解决上面遇到的读写问题！