Article Directory
1.Redis master-slave replication
1.1 Overview
- Redis, although the speed of reading and writing is very fast, but it will also produce a situation where the reading pressure is very high. In order to share the read pressure, Redis supports master-slave replication to ensure that the data content of the master database is completely consistent with the content of the slave database.
1.2 Redis structure classification
- The master-slave structure of Redis can adopt a master-multi-slave or cascade structure,
- Redis master-slave replication can be divided into full synchronization and incremental synchronization according to whether it is full.
1.2.1 Redis full synchronization
- Redis full replication generally occurs during the initialization phase of the slave. At this time, the slave needs to copy all the data on the master.
- Specific steps are as follows:
从服务器连接主服务器,发送SYNC命令;
主服务器接收到sYNC命名后,开始执行BGSAVE命令生成RDB文件并使用缓冲区记录此后执行的所有写命令;
主服务器BGSAVE执行完后,向所有从服务器发送快照文件,并在发送期间继续记录被执行的写命令;
从服务器收到快照文件后丢弃所有旧数据,载入收到的快照;
主服务器快照发送完毕后开始向从服务器发送缓冲区中的写命令;
从服务器完成对快照的载入,开始接收命令请求,并执行来自主服务器缓冲区的写命令。
1.2.2 Redis incremental replication
-
Redis incremental replication refers to the process of synchronizing write operations from the master server to the slave server when the slave is initialized and starts to work normally. ·
-
The process of incremental replication is mainly that the master server sends the same write command to the slave server every time it executes a write command, and the slave server receives and executes the received write command.
1.3 master-slave synchronization strategy
- When the master and slave are just connected, perform full synchronization;
- After the full synchronization is over, perform incremental synchronization;
- Of course, if necessary, the slave can initiate full synchronization at any time. The redis strategy is to try incremental synchronization first anyway; if unsuccessful, the slave is required to perform full synchronization.
2. Build Redis master-slave replication
- Project environment
一台mater主服务器
IP:192.168.140.20
两台slave备选服务器
IP:192.168.140.21
IP:192.168.140.22
2.1 Open the redis_server service on the server
Note: The redis service needs to be turned on on the primary and standby servers
发送所有会话,关闭防火墙
systemctl stop firewalld
setenforce 0
- Import the following software packages on all three servers and place them in the root directory
2.1.1 Unzip and install
[root@master ~]# tar zxvf redis-5.0.4.tar.gz
[root@master ~]# cd redis-5.0.4/
[root@master redis-5.0.4]# make
[root@master redis-5.0.4]# make PREFIX=/usr/local/redis install
[root@master redis-5.0.4]# cd
2.1.2 Create a link file and start the service
[root@master ~]# ln -s /usr/local/redis/bin/* /usr/local/bin/
[root@master ~]# cd redis-5.0.4/utils/
[root@master utils]# ./ //按table 查看相关命令
create-cluster/ graphs/ hyperloglog/ lru/ redis_init_script.tpl speed-regression.tcl
generate-command-help.rb hashtable/ install_server.sh redis_init_script releasetools/ whatisdoing.sh
[root@master utils]# ./install_server.sh
...//弹出的信息,按回车即可
[root@master utils]# netstat -anpt | grep redis
tcp 0 0 127.0.0.1:6379 0.0.0.0:* LISTEN 60219/redis-server
2.2 Modify the configuration file
- On the main server
[root@master ~]# vi /etc/redis/6379.conf
...
69 bind 0.0.0.0 //修改监听地址为 0.0.0.0 (在实验环境使用),现网环境建议绑定从服务器IP地址
136 daemonize yes //开启守护进程
171 logfile /var/log/redis_6379.log //修改日志文件目录
263 dir /var/lib/redis/6379 //修改工作目录
699 appendonly yes //开启AOF持久化功能
- On the alternate server, first enable the same functions as the main server, and then perform the following configuration
[root@slave1 ~]# vi /etc/redis/6379.conf
...
replicaof 192.168.140.20 6379 //开启复制功能
[root@slave1 ~]# /etc/init.d/redis_6379 restart
Stopping ...
Waiting for Redis to shutdown ...
Redis stopped
Starting Redis server...
- View logs on the main server
2.2 Verify the slave node on the master server
- Connect to the database and edit content on the main server
[root@master ~]# redis-cli //连接数据库
127.0.0.1:6379> info replication //查看节点信息
# Replication
role:master
connected_slaves:2
slave0:ip=192.168.140.21,port=6379,state=online,offset=448,lag=1
slave1:ip=192.168.140.22,port=6379,state=online,offset=448,lag=1
master_replid:8e6ded5a687920d545a8133c8c0093d8e962056d
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:448
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:448
127.0.0.1:6379> set cp 9 //写入内容,并验证
OK
127.0.0.1:6379> get cp
"9"
- Verify the edited content on the master server from the server
[root@slave1 ~]# redis-cli
127.0.0.1:6379> get cp
"9"
127.0.0.1:6379>
2.3 Verification results
- When the master node goes down
[root@master ~]# /etc/init.d/redis_6379 stop
Stopping ...
Redis stopped
(1) View the log on the slave node
(2) View on the server whether the edited content on the master server exists
[root@slave1 ~]# redis-cli
127.0.0.1:6379> get cp
"9"
127.0.0.1:6379>
-
At this time, it means that the redis master-slave is successfully built
-
Conclusion: Redis master-slave replication, when the master server fails, the standby server will not be automatically switched. Although reading data is normal, writing data will cause problems.
3. Sentry mode
注意:哨兵模式,一定是基于redis主从复制模式下做的
3.1 Principles of Sentinel Mode
- Sentinel is a distributed system used to monitor each server in the master-slave structure. When a failure occurs, a new master is selected through a voting mechanism and all slaves are connected to the new master;
- Therefore, the number of the entire cluster running sentinels must not be less than 3 nodes.
3.2 The role of sentinel mode
-
Monitoring
Continuously check whether master and slave are running normally;
master survival detection, master and slave running status detection. -
Notification (Reminder)
When the monitored server has a problem, send a notification to other (sentinel room, client). -
Automatic failover
Disconnect the master and slave, select one slave as the master, connect other slaves to the new master, and inform the client of the new server address
Supplement: Sentinel is also a redis server, but does not provide data services
3.3 Start of Sentry Mode
- The startup of the sentry depends on the master-slave mode, so the sentry mode must be installed when the master-slave mode is installed. The sentry mode needs to be deployed on all nodes;
- The sentinel mode will monitor whether all redis working nodes are normal. When the master has a problem, because other nodes lose contact with the master node, it will vote. After the vote is half of the vote, it is considered that the master does have a problem, and then the sentry room will be notified, and then from Select one of the slaves as the new master
3.4 Sentinel configuration
3.4.1 Edit configuration file
[root@master ~]# vi redis-5.0.4/sentinel.conf
17 protected-mode no //关闭保护模式
26 daemonize yes //指定sentinel为后台启动(开启守护进程)
36 logfile "/var/log/sentinel.log" //指定日志存放路径
65 dir /var/lib/redis/6379 //指定数据库存放路径
84 sentinel monitor mymaster 192.168.140.20 6379 2
//至少几个哨兵检测到主服务器故障了,才会进行故障迁移(当主服务器故障,切换到slave为主服务器时,会将地址改为一个ID号)
113 sentinel down-after-milliseconds mymaster 3000 //判定服务器down掉的时间周期,默认30000毫秒(30S)
146 sentinel failover-timeout mymaster 180000 //故障节点的最大超时时间为180000(180秒)
3.4.2 Copy configuration file to slave server
[root@master ~]# scp redis-5.0.4/sentinel.conf root@192.168.140.21:/root/redis-5.0.4
3.4.3 Start Sentry Mode
先启动master,再启slave
[root@master ~]# redis-sentinel redis-5.0.4/sentinel.conf & //&表示在后台启动
[1] 63649
[root@slave1 ~]# redis-sentinel redis-5.0.4/sentinel.conf &
[1] 62327
[root@slave2 ~]# redis-sentinel redis-5.0.4/sentinel.conf &
[1] 60481
3.4.4 View sentinel information
- View log
[root@master ~]# tail -f /var/log/sentinel.log
- View process
[root@master ~]# ps aux | grep sentinel
root 63650 0.2 0.1 153836 2668 ? Ssl 17:23 0:01 redis-sentinel *:26379 [sentinel]
root 63739 0.0 0.0 112676 980 pts/1 S+ 17:31 0:00 grep --color=auto sentinel
[root@master ~]# ps aux | grep redis
root 62860 0.1 0.1 156396 2800 ? Ssl 16:03 0:06 /usr/local/bin/redis-server 0.0.0.0:6379
root 63650 0.2 0.1 153836 2668 ? Ssl 17:23 0:01 redis-sentinel *:26379 [sentinel]
root 63742 0.0 0.0 112676 984 pts/1 S+ 17:31 0:00 grep --color=auto redis
- View sentry status
[root@master ~]# redis-cli -h 192.168.140.20 -p 26379 info sentinel
# 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=192.168.140.20:6379,slaves=2,sentinels=3
[root@master ~]# redis-cli -h 192.168.140.20 -p 6379 info replication
# Replication
role:master
connected_slaves:2
slave0:ip=192.168.140.21,port=6379,state=online,offset=174385,lag=0
slave1:ip=192.168.140.22,port=6379,state=online,offset=174385,lag=0
master_replid:75cd6e32461f20e72b1ae7a09bffff02803fb9f9
master_replid2:0000000000000000000000000000000000000000
master_repl_offset:174385
second_repl_offset:-1
repl_backlog_active:1
repl_backlog_size:1048576
repl_backlog_first_byte_offset:1
repl_backlog_histlen:174385
3.5 Simulate failure
- Stop the redis service on the main server
[root@master ~]# /etc/init.d/redis_6379 stop //停止redis服务
也可以用 kill 命令
kill -9 进程号
3.6 Verification results
- View the log on the slave server
[root@slave1 ~]# tail -f /var/log/sentinel.log
[root@slave2 ~]# tail -f /var/log/sentinel.log
[root@slave2 ~]# redis-cli -h 192.168.140.22 -p 26379 info sentinel
# 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=192.168.140.22:6379,slaves=2,sentinels=3
- When the master server master restarts,
check the log and the sentinel status and find that the master does not preempt the slave server
[root@master ~]# /etc/init.d/redis_6379 start