Shulou(Shulou.com)05/31 Report--

This article mainly introduces the relevant knowledge of "how to configure Redis cluster master-slave mode". The editor shows you the operation process through an actual case. The operation method is simple, fast and practical. I hope this article "how to configure Redis cluster master-slave mode" can help you solve the problem.

First, why do you need clusters?

In our actual development, it is not possible to use only one Redis in an engineering project for the following reasons:

(1) structurally, a single Redis server will have a single point of failure, and a server needs to handle all the request load, so there is a lot of pressure.

(2) in terms of capacity, the memory capacity of a single Redis server is limited. Even if the memory capacity of a Redis server is 256g, you cannot use all the memory as Redis storage memory. Generally speaking, the maximum memory used by a single Redis should not exceed 20g.

(3) the read and write performance of a single Redis server is limited, and the read and write ability can be improved by using cluster.

Second, the introduction of master-slave mode

At present, Redis has three cluster modes, namely: master-slave mode, Sentinel mode and Cluster mode. Master-slave mode is the simplest of the three modes. In master-slave replication, it refers to copying the data of one Redis server to other Redis servers. The former is called the master node (master/leader) and the latter is called the slave/follower node.

Note:

(1) the replication of data is one-way and can only be from master node to slave node. Master is mainly written, while Slave is mainly read.

(2) by default, each Redis server is the primary node

(3) A master node can have multiple slave nodes (or no slave nodes), but a slave node can only have one master node.

Action

1. Data redundancy: master-slave replication realizes the hot backup of data, which is a way of data redundancy in addition to persistence.

2. Fault recovery: when there is a problem with the master node, the service can be provided by the slave node to achieve rapid fault recovery; in fact, it is a kind of service redundancy.

3. High availability (cluster) cornerstone: master-slave replication or the basis on which sentinels and clusters can be implemented, so master-slave replication is the basis of high availability for Redis.

4. Load balancing: on the basis of master-slave replication, with the separation of reads and writes, the master node can provide write services, and the slave nodes can provide read services (that is, the master node is connected to the master node when writing Redis data, and the slave node is connected to the slave node when reading Redis data) to share the server load. Especially in the scenario of writing less and reading more, the concurrency of the Redis server can be greatly improved by sharing the read load among multiple slave nodes.

For example, on our e-commerce website, we can find that for a product that only needs to be uploaded once, but it can be viewed many times by users, that is, in the case of "write less and read more", we can use master-slave replication to separate reading and writing. ease the pressure on the server:

Third, build master-slave cluster 3.1 and prepare for the work.

1. Copy three configuration files (original name: redis.conf) and rename them to: redis79.conf,redis80.conf,redis81.conf.

2. Modify the configuration file

(1) modify redis79.conf

Modify port number

Port 6379

Set to run in the background

Daemonize:yes

Set the name of the log file

Logfile "6379.log"

Set the db file name

Dbfilename dump6379.rdb

(2) modify redis80.conf

Modify port number

Port 6380

Set to run in the background

Daemonize:yes

Set the recording process Id file name

Pidfile / var/run/redis_6380.pid

Set the name of the log file

Logfile "6380.log"

Set the db file name

Dbfilename dump6380.rdb

(3) modify redis81.conf

Modify port number

Port 6381

Set to run in the background

Daemonize:yes

Set the recording process Id file name

Pidfile / var/run/redis_6381.pid

Set the name of the log file

Logfile "6381.log"

Set the db file name

Dbfilename dump6381.rdb

The effects of these attributes are as follows:

Pid (port ID): the ID of the process is recorded and the file is locked. It can prevent multiple starts of the program.

Logfile: identify the location of the log file

Dbfilename:dumpxxx.file # persistent file location

Port: the port number occupied by the process

Set up one master and two slaves to start the Redis server

Note: by default, each Reids server is the master node, and we only need to build the master and slave in the slave machine.

Now start the redis79,redis80,redis81 server separately.

Redis-server redis79.conf redis-server redis80.confredis-server redis81.conf

Use the following command to see if the startup was successful:

Ps-ef | grep redis

Open three client windows, each operating three Redis servers.

Enter the command:

Note that you need to specify the port so that you know which Redis we want to open.

Window one:

Redis-cli-p 6379

Window 2:

Redis-cli-p 6380

Window 3:

Redis-cli-p 6381 sets the master-slave relationship

We set redis79 as the master node and redis80 and redis81 as slave nodes.

Configuring the IP address and port number of the host is equivalent to wanting to recognize it as your boss.

Redis80:

# SLAVEOF IP address port 127.0.0.1 6379OK > slaveof 127.0.0.1 6379OK

Redis81:

# SLAVEOF IP address port 127.0.0.1 6379OK > slaveof 127.0.0.1 6379OK

At this point, we can check the master-slave relationship by using the INFO command on the slave:

Info replication

At this point, we go to the host redis79 and use the same command to check:

Now our one-master-two-slave relationship has been successfully built!

Tip: if we want to turn the slave into a host, we only need to execute the following command on the slave to make ourselves a host.

SLAVEOF no one IV. Knowledge explanation knowledge 1.

The host can read and write, while the slave can only read.

Note: all information and data in the host will be automatically saved from the computer.

Host:

127.0.1 6379 > set key1 v1OK127.0.0.1:6379 > get key1 "v1"

Slave:

127.0.0.1error 6380 > get key1 "v1" 127.0.0.1 error 6380 > set key2 v2 # will report an error when writing, indicating that the slave machine can only perform read operation (read operation) READONLY You can't write against a read only replica. Knowledge two

If the host is down, the slave can still read the data before the downtime, but there is still no write operation. If the host is restored, the slave can still get the data written by the host.

(1) stop the host process (demonstrate that the host is down)

Command to stop the process:

Kill-9 pid # pid is the redis process number

(2) obtain the data written by the host before downtime from the computer

It can be found that you can get it smoothly, but it is still impossible to write.

(3) restore the host

Redis-server redis79.conf

(4) the host rewrites the data and obtains the latest data from the computer.

Host writes data:

127.0.0.1 6379 > set K2 yixinOK

Read the latest data from the machine:

127.0.1 6380 > get K2 "yixin" knowledge III

Disconnection of slave machine under two configuration modes

A. set the master-slave relationship on the command line

When the slave machine is disconnected, it becomes the host after reconnecting, and you can get the data before the disconnection, but you can't get the value newly written by the host. If you reset the master-slave relationship, you can get all the data of the host.

(1) stop the slave process.

(2) the host writes new data.

127.0.0.1 6379 > set K3 newOK

(3) restart the slave server.

Redis-server redis80.conf

(4) try to obtain the data written by the host before the outage, and it is found that it can be obtained.

127.0.0.1 6380 > get K1 "v1"

(5) try to get the data written by the host during the outage, but it is found that it cannot be obtained.

127.0.0.1 6380 > get K3 (nil)

This time we can check the master-slave relationship, because it is configured on the command line, so it changes back to the host after reboot.

127.0.0.1 6380 > info replication# Replicationrole:masterconnected_slaves:0

(6) if you want to get all the data of the host, just execute the following command to reconfigure the master-slave relationship.

Slaveof 127.0.0.1 6379b, master-slave relationship of profile settings

After disconnecting from the machine, reconnect, you can also get all the data of the host.

(1) modify the configuration file redis80.conf to add a master-slave relationship.

# specify the ip and portslaveof of the host 127.0.0.1 6379

(2) add new data to the host

127.0.0.1 6379 > set K5 helloOK

(3) restart redis80 server.

Redis-server redis80.conf

(4) obtain the newly written data from the host during the downtime and find that it can be obtained smoothly now.

127.0.0.1 6380 > get K5 "hello"

If we look at the master-slave relationship of 6380, we can see that the master-slave relationship has been set up at the time of restart.

The principle of replication

(1) when Slave starts to connect to Master successfully, it will send a sync synchronization command.

(2) Master receives the order to start the save process in the background, and collects all the commands received to modify the dataset. After the background process is completed, master will transfer the whole data file to slave and complete a complete synchronization.

(3) full copy: after receiving the database file data, the slave service saves it and loads it into memory.

(4) incremental copy: Master continues to send all new collected modification commands to slave in turn to complete the synchronization.

Note: as long as you reconnect to master, a full synchronization (full copy) will be performed automatically! Our data must be seen in the slave computer.

VI. Advantages and disadvantages of master-slave mode

(1) multiple Slaves can be synchronized with the same Master.

(2) Slave can also accept connection and synchronization requests from other Slaves, which can effectively offload the synchronization pressure of Master. So we can think of Redis's Replication architecture as a graph structure.

(3) Master Server provides services for Slaves in a non-blocking way. So during Master-Slave synchronization, the client can still submit queries or modify requests.

(4) Slave Server also completes data synchronization in a non-blocking way. During synchronization, if a client submits a query request, Redis returns the data before synchronization.

(5) in order to offload the read operation pressure of Master, the Slave server can provide read-only service for the client, and the write service must still be completed by Master. Even so, the scalability of the system has been greatly improved.

(6) Master can leave the data saving operation to Slaves to complete, thus avoiding the need for an independent process in Master to complete this operation.

(7) Master-slave replication is supported, and the host will automatically synchronize the data to the slave, and read and write can be separated.

Shortcoming

(1) Redis master-slave mode does not have automatic fault tolerance and recovery. If the master node goes down, the Redis cluster will not work. Human intervention is needed to promote the slave node to the master node.

(2) if a part of the data is not synchronized to the slave in time before the host goes down, it will cause data inconsistency even after switching the host, thus reducing the availability of the system.

(3) because there is only one master node, its writing capacity and storage capacity are limited to a certain extent.

(4) in the total data synchronization, the phenomenon of stutter may be caused if the amount of synchronized data is large.

This is the end of the content about "how to configure the master-slave mode of Redis cluster". Thank you for your reading. If you want to know more about the industry, you can follow the industry information channel. The editor will update different knowledge points for you every day.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.