Redis 哨兵模式（windows）

https://blog.csdn.net/weixin_41846320/article/details/83753182

 

1、以文章为基础：主从模式

2、分别在三个redis目录下新增  sentinel.conf  文件：

port 26379
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13

port 26479
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13

port 26579
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13

3、增加哨兵启动脚本  startup_s.bat（其他类似）关键字是必须--sentinel：

title slaver_6381_s
redis-server.exe sentinel.conf --sentinel

4、总启动脚本：

cd master_6379
start "master_6379" "F:\MiddleWares\master_6379\startup.bat"
start "master_6379_s" "F:\MiddleWares\master_6379\startup_s.bat"
 
cd ../
cd slaver_6380
start "slaver_6380" "F:\MiddleWares\slaver_6380\startup.bat"
start "slaver_6380_s" "F:\MiddleWares\slaver_6380\startup_s.bat"
 
cd ../
cd slaver_6381
start "slaver_6381" "F:\MiddleWares\slaver_6381\startup.bat"
start "slaver_6381_s" "F:\MiddleWares\slaver_6381\startup_s.bat"

5、启动结果：

启动后  sentinel.conf 文件的变化（6379~6381）：

port 26379
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\master_6379"
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-slave mymaster 127.0.0.1 6380
sentinel known-sentinel mymaster 127.0.0.1 26579 81435f5e909f96330cf51e33db89f0908a9a669e
sentinel known-sentinel mymaster 127.0.0.1 26479 9a146d63a04fceb6630b95fd1eab6d2492011531
sentinel current-epoch 12

port 26479
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\slaver_6380"
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-slave mymaster 127.0.0.1 6380
sentinel known-sentinel mymaster 127.0.0.1 26579 81435f5e909f96330cf51e33db89f0908a9a669e
sentinel known-sentinel mymaster 127.0.0.1 26379 ff5e921d1d156ff53664ec4963930a1681306c7d
sentinel current-epoch 12

port 26579
sentinel monitor mymaster 127.0.0.1 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 12
sentinel leader-epoch mymaster 13
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\slaver_6381"
sentinel known-slave mymaster 127.0.0.1 6380
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-sentinel mymaster 127.0.0.1 26479 9a146d63a04fceb6630b95fd1eab6d2492011531
sentinel known-sentinel mymaster 127.0.0.1 26379 ff5e921d1d156ff53664ec4963930a1681306c7d
sentinel current-epoch 12

6、连接测试：

Jedis jedis = new Jedis("localhost", 6379);
System.out.println(jedis.info("Replication"));
jedis = new Jedis("localhost", 6380);
System.out.println(jedis.info("Replication"));
jedis = new Jedis("localhost", 6381);
System.out.println(jedis.info("Replication"));
jedis = new Jedis("localhost", 26379);
System.out.println(jedis.info("sentinel"));
jedis = new Jedis("localhost", 26479);
System.out.println(jedis.info("sentinel"));
jedis = new Jedis("localhost", 26579);
System.out.println(jedis.info("sentinel"));

# Replication
role:master
connected_slaves:2
slave0:ip=127.0.0.1,port=6380,state=online,offset=265271,lag=0
slave1:ip=127.0.0.1,port=6381,state=online,offset=265420,lag=0
master_repl_offset:265555
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:2
repl_backlog_histlen:265554

# Replication
role:slave
master_host:127.0.0.1
master_port:6379
master_link_status:up
master_last_io_seconds_ago:0
master_sync_in_progress:0
slave_repl_offset:265555
slave_priority:100
slave_read_only:1
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

# Replication
role:slave
master_host:127.0.0.1
master_port:6379
master_link_status:up
master_last_io_seconds_ago:0
master_sync_in_progress:0
slave_repl_offset:265555
slave_priority:100
slave_read_only:1
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6379,slaves=2,sentinels=3

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6379,slaves=2,sentinels=3

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6379,slaves=2,sentinels=3

7、同样，从机是只读

8、关掉master之后的连接测试（6379已经connect refused）：

Jedis jedis = new Jedis("localhost", 6380);
System.out.println(jedis.info("Replication"));
jedis = new Jedis("localhost", 6381);
System.out.println(jedis.info("Replication"));
jedis = new Jedis("localhost", 26379);
System.out.println(jedis.info("sentinel"));
jedis = new Jedis("localhost", 26479);
System.out.println(jedis.info("sentinel"));
jedis = new Jedis("localhost", 26579);
System.out.println(jedis.info("sentinel"));

# Replication
role:master
connected_slaves:1
slave0:ip=127.0.0.1,port=6381,state=online,offset=25451,lag=1
master_repl_offset:25586
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:2
repl_backlog_histlen:25585

# Replication
role:slave
master_host:127.0.0.1
master_port:6380
master_link_status:up
master_last_io_seconds_ago:0
master_sync_in_progress:0
slave_repl_offset:25586
slave_priority:100
slave_read_only:1
connected_slaves:0
master_repl_offset:0
repl_backlog_active:0
repl_backlog_size:1048576
repl_backlog_first_byte_offset:0
repl_backlog_histlen:0

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6380,slaves=2,sentinels=3

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6380,slaves=2,sentinels=3

# Sentinel
sentinel_masters:1
sentinel_tilt:0
sentinel_running_scripts:0
sentinel_scripts_queue_length:0
master0:name=mymaster,status=ok,address=127.0.0.1:6380,slaves=2,sentinels=3

sentinel.conf  配置文件的变化（6379~6381）：

port 26379
sentinel monitor mymaster 127.0.0.1 6380 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 14
sentinel leader-epoch mymaster 14
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\master_6379"
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-slave mymaster 127.0.0.1 6379
sentinel known-sentinel mymaster 127.0.0.1 26579 81435f5e909f96330cf51e33db89f0908a9a669e
sentinel known-sentinel mymaster 127.0.0.1 26479 9a146d63a04fceb6630b95fd1eab6d2492011531
sentinel current-epoch 14

port 26479
sentinel monitor mymaster 127.0.0.1 6380 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 14
sentinel leader-epoch mymaster 14
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\slaver_6380"
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-slave mymaster 127.0.0.1 6379
sentinel known-sentinel mymaster 127.0.0.1 26579 81435f5e909f96330cf51e33db89f0908a9a669e
sentinel known-sentinel mymaster 127.0.0.1 26379 ff5e921d1d156ff53664ec4963930a1681306c7d
sentinel current-epoch 14

port 26579
sentinel monitor mymaster 127.0.0.1 6380 1
sentinel down-after-milliseconds mymaster 5000
sentinel config-epoch mymaster 14
sentinel leader-epoch mymaster 14
# Generated by CONFIG REWRITE
dir "F:\\MiddleWares\\slaver_6381"
sentinel known-slave mymaster 127.0.0.1 6379
sentinel known-slave mymaster 127.0.0.1 6381
sentinel known-sentinel mymaster 127.0.0.1 26479 9a146d63a04fceb6630b95fd1eab6d2492011531
sentinel known-sentinel mymaster 127.0.0.1 26379 ff5e921d1d156ff53664ec4963930a1681306c7d
sentinel current-epoch 14

• 6380 的 redis.windows.conf  配置文件中的  "slaveof 127.0.0.1 6379"  被删除
• 6381 的 redis.windows.conf  配置文件中的  "slaveof 127.0.0.1 6379"  被改为 "slaveof 127.0.0.1 6380"
• 也就是说6380成为了主节点，并且拥有之前6379主节点的功能

9、如果关掉了master 的 sentinel 之后：

连接测试中查看 26479  和  26579 ，sentinel的值仍然显示为3：

master0:name=mymaster,status=ok,address=127.0.0.1:6379,slaves=2,sentinels=3

并且再关闭master之后，6380 节点仍然可以变更为主节点，只是原有的 6379节点中的 sentinel.conf  文件不会被修改（6380和6381的会）；

10、关于 sentinel.conf 配置文件的解释：

#当前Sentinel服务运行的端口
port 26379

#Sentinel去监视一个名为mymaster的主redis实例，这个主实例的IP地址为本机地址127.0.0.1，端口号为6379，
#而将这个主实例判断为失效至少需要2个 Sentinel进程的同意，只要同意Sentinel的数量不达标，自动failover就不会执行
sentinel monitor mymaster 127.0.0.1 6379 1

#指定了Sentinel认为Redis实例已经失效所需的毫秒数。当 实例超过该时间没有返回PING，或者直接返回错误，那么Sentinel将这个实例标记为主观下线。
#只有一个 Sentinel进程将实例标记为主观下线并不一定会引起实例的自动故障迁移：只有在足够数量的Sentinel都将一个实例标记为主观下线之后，实例才会被标记为客观下线，这时自动故障迁移才会执行
sentinel down-after-milliseconds mymaster 5000

#指定了在执行故障转移时，最多可以有多少个从Redis实例在同步新的主实例，在从Redis实例较多的情况下这个数字越小，同步的时间越长，完成故障转移所需的时间就越长
sentinel config-epoch mymaster 12

#如果在该时间（ms）内未能完成failover操作，则认为该failover失败
sentinel leader-epoch mymaster 13