Redis集群-哨兵模式搭建（1主2从3哨兵节点）

原创北极星运维记事 2023-04-20 22:47 发表于四川

收录于合集#redis8个

主机规划

类型	IP地址	端口号
主	192.168.77.145	6379
从1	192.168.77.146	6379
从2	192.168.77.147	6379
哨兵1	192.168.77.145	26379
哨兵2	192.168.77.146	26379
哨兵3	192.168.77.147	26379

1主2从搭建

见主从复制搭建，Redis集群-主从复制搭建

哨兵配置

哨兵配置每个节点执行。

从安装包中，复制哨兵配置文件到redis配置中，作为哨兵节点的配置文件

 cp /opt/redis-7.0.4/sentinel.conf /opt/redis/conf/

编辑哨兵配置文件修改内容如下：

port 26379 #设置为后台启动 daemonize yes pidfile "/opt/redis/tmp/redis-sentinel.pid" logfile "/opt/redis/logs/redis-sentinel.log" dir "/opt/redis" #哨兵sentinel监控的redis主节点的 ip port  sentinel monitor mymaster 192.168.77.145 6379 2 #当在Redis实例中开启了requirepass，所有连接Redis实例的客户端都要提供密码。需要注意的是，master实例也需要配置masterauth，否则master实例下线重新上线后，会无法加入到集群中 sentinel auth-pass mymaster 123456

启动哨兵服务

 /opt/redis/bin/redis-sentinel /opt/redis/conf/sentinel.conf

哨兵配置查看

 [root@dba Wed Apr 19 09:38 ~]# redis-cli -p 26379 info Sentinel # Sentinel sentinel_masters:1 sentinel_tilt:0 sentinel_tilt_since_seconds:-1 sentinel_running_scripts:0 sentinel_scripts_queue_length:0 sentinel_simulate_failure_flags:0 master0:name=mymaster,status=ok,address=192.168.77.145:6379,slaves=2,sentinels=3

模拟故障转移

关闭主节点192.168.77.145:6379,间隔一会后再次查看可以看到master节点变为了192.168.77.147:6379

[root@dba Wed Apr 19 09:40 ~]# redis-cli -a 123456 shutdown [root@dba Wed Apr 19 09:42 ~]# redis-cli -p 26379 info Sentinel # Sentinel sentinel_masters:1 sentinel_tilt:0 sentinel_tilt_since_seconds:-1 sentinel_running_scripts:0 sentinel_scripts_queue_length:0 sentinel_simulate_failure_flags:0 master0:name=mymaster,status=ok,address=192.168.77.147:6379,slaves=2,sentinels=3

在选举后的master节点192.168.77.147:6379上查看当前仅有192.168.77.146:6379节点在线，192.168.77.145:6379因为已经shutdown了

 [root@dbc ~]# redis-cli -a 123456 info replication # Replication role:master connected_slaves:1 slave0:ip=192.168.77.146,port=6379,state=online,offset=1114835,lag=1 master_failover_state:no-failover master_replid:c556a04bd19dfc7627743569c9f2613dc8a7232b master_replid2:ce16d352e923b9c77de6dd925f289dcd27e590ca master_repl_offset:1115121 second_repl_offset:1090349 repl_backlog_active:1 repl_backlog_size:1048576 repl_backlog_first_byte_offset:1051856 repl_backlog_histlen:63266

启动恢复192.168.77.145:6379节点服务，并再次在检查发现之前shutdown掉的主节点192.168.77.145:6379变为了从节点了加入了当前集群。

 [root@dbc ~]# redis-cli -a 123456 info replication # Replication role:master connected_slaves:2 slave0:ip=192.168.77.146,port=6379,state=online,offset=1147649,lag=0 slave1:ip=192.168.77.145,port=6379,state=online,offset=1147792,lag=0 master_failover_state:no-failover master_replid:c556a04bd19dfc7627743569c9f2613dc8a7232b master_replid2:ce16d352e923b9c77de6dd925f289dcd27e590ca master_repl_offset:1147958 second_repl_offset:1090349 repl_backlog_active:1 repl_backlog_size:1048576 repl_backlog_first_byte_offset:1051856 repl_backlog_histlen:96103