redis主从架构与redis+sentinel 哨兵机制架

Replication

role:master
connected_slaves:1

在从redis上可以看到
[root@prd3-redis02-10-184 conf]# redis-cli -aXXX
Warning: Using a password with '-a' option on the command line interface may not be safe.
127.0.0.1:6379> info
####### Replication
role:slave
master_host:192.168.10.183
master_port:6379

这样redis的主从就搭建好了

搭建完redis后，目录结构

[root@prd3-redis01-10-183 ivargo]# pwd /ivargo [root@prd3-redis01-10-183 ivargo]# tree . ├── app │ ├── redis │ │ ├── 6379.pid │ │ ├── conf │ │ │ └── 6379.conf │ │ ├── data │ │ │ ├── 6379.rdb │ │ │ └── appendonly_6379.aof │ │ └── log │ └── sentinel │ └── conf └── log └── 6379.log 8 directories, 5 files

实现redis的sentinel [root@prd3-redis01-10-183 conf]# pwd /ivargo/app/sentinel/conf [root@prd3-redis01-10-183 conf]# cat sentinel.conf //这个配置文件是prd3上的sentinel的配置文件，我们需要修改的 # Example sentinel.conf # bind 127.0.0.1 192.168.1.1 bind 0.0.0.0 protected-mode no port 26379 daemonize yes loglevel notice logfile "/ivargo/log/sentinel.log" dir "/tmp" # sentinel auth-pass <master-name> <password> # sentinel down-after-milliseconds <master-name> <milliseconds> # Default is 30 seconds. sentinel myid ba794535d3a65e14b266e28462fa7c68de609f39 # sentinel parallel-syncs <master-name> <numslaves> sentinel deny-scripts-reconfig yes # sentinel failover-timeout <master-name> <milliseconds> # Default is 3 minutes = 180000. sentinel monitor mymaster 10.80.85.20 6379 2 # Generated by CONFIG REWRITE sentinel down-after-milliseconds mymaster 6000 sentinel auth-pass mymaster xxx sentinel config-epoch mymaster 0 sentinel leader-epoch mymaster 1 sentinel known-slave mymaster 10.80.85.21 6379 sentinel known-sentinel mymaster 10.80.85.21 26379 c3002e02c6a2d0041afd2569f3fd1985bc38993e sentinel current-epoch 1

[root@prd3-redis01-10-183 conf]# cat sentinel.conf //这里面都是我们自己指定的，他其他的文件内容，都是他自己生成的 # Example sentinel.conf # bind 127.0.0.1 192.168.1.1 bind 0.0.0.0 protected-mode no port 26379 daemonize yes loglevel notice logfile "/ivargo/log/sentinel.log" pidfile "/ivargo/app/sentinel/26379.pid" dir "/tmp" sentinel monitor mymaster 192.168.10.183 6379 2 sentinel auth-pass mymaster XXX 184上的sentinel配置文件和183一样的 [root@prd3-redis01-10-183 conf]# redis-sentinel /ivargo/app/sentinel/conf/sentinel.conf [root@prd3-redis02-10-184 conf]# redis-sentinel /ivargo/app/sentinel/conf/sentinel.conf

[root@prd3-redis01-10-183 conf]# cat sentinel.conf # Example sentinel.conf # bind 127.0.0.1 192.168.1.1 bind 0.0.0.0 protected-mode no port 26379 daemonize yes loglevel notice logfile "/ivargo/log/sentinel.log" pidfile "/ivargo/app/sentinel/26379.pid" dir "/tmp" sentinel myid 6075d58bdc7c2602f98d90c0aa48470c4dbb50d9 sentinel deny-scripts-reconfig yes # Generated by CONFIG REWRITE sentinel monitor mymaster 192.168.10.183 6379 2 sentinel auth-pass mymaster XXX sentinel config-epoch mymaster 0 sentinel leader-epoch mymaster 0 sentinel known-slave mymaster 192.168.10.184 6379 sentinel known-sentinel mymaster 192.168.10.184 26379 039bd11a572245b6c16c6e204523d781ffb6ba4c sentinel current-epoch 0 查看不同点我们检查 [root@prd3-redis01-10-183 conf]# redis-cli -h 192.168.10.183 -p 26379 info # Sentinel sentinel_masters:1 sentinel_tilt:0 sentinel_running_scripts:0 sentinel_scripts_queue_length:0 sentinel_simulate_failure_flags:0 master0:name=mymaster,status=ok,address=192.168.10.183:6379,slaves=1,sentinels=2

验证sentinel高可用 [root@prd3-redis01-10-183 redis]# kill `cat 6379.pid` 杀死183的进程然后在184上看 [root@prd3-redis02-10-184 conf]# redis-cli -h 192.168.10.183 -p 26379 info # Sentinel sentinel_masters:1 sentinel_tilt:0 sentinel_running_scripts:0 sentinel_scripts_queue_length:0 sentinel_simulate_failure_flags:0 master0:name=mymaster,status=ok,address=192.168.10.184:6379,slaves=1,sentinels=2 //主节点变成了184了在启动183 [root@prd3-redis01-10-183 redis]# redis-server /ivargo/app/redis/conf/6379.conf [root@prd3-redis01-10-183 redis]# redis-cli -h 192.168.10.183 -p 26379 info # Sentinel sentinel_masters:1 sentinel_tilt:0 sentinel_running_scripts:0 sentinel_scripts_queue_length:0 sentinel_simulate_failure_flags:0 master0:name=mymaster,status=ok,address=192.168.10.184:6379,slaves=1,sentinels=2 然后查看redis 原来的183是主节点，现在是从节点了 [root@prd3-redis01-10-183 redis]# redis-cli -h 192.168.10.183 -p 6379 -a XXX info # Replication role:slave master_host:192.168.10.184 master_port:6379 [root@prd3-redis02-10-184 conf]# redis-cli -h 192.168.10.184 -p 6379 -a XXX info # Replication role:master connected_slaves:1 slave0:ip=192.168.10.183,port=6379,state=online,offset=169470,lag=1 以上实现了redis的sentinel机制也就是 HA机制

其实对于现上环境的话，当redis发生了主从切换的时候，这个比如程序调用的ip就会改变，写两个ip和两个sentinel地址显然是不合理的，我们可以实现vip机制自动切换，来实现程序调用只需要一个ip，相关内容，请期待。

Replication

相关阅读