Explain the principle of MySQL master-slave replication in detail

Hello everyone , I meet you again , I'm the king of the whole stack .

brief introduction ： What is master-slave replication , How to realize the separation of reading and writing , Read this article and you will understand ！

Mind mapping

Preface

In many projects , Especially Internet projects , In the use of MySQL Master slave replication will be used 、 Read write separation architecture .

Why should we adopt a master-slave replication read-write separation architecture ？ How to achieve ？ What's the downside ？ Let's start this learning journey with these questions ！

Why use master-slave replication 、 Read / write separation

Master slave copy 、 Read write separation is usually used together . The goal is simple , Just for Improve the concurrent performance of database . Would you , Suppose it's a single machine , Reading and writing are all on the same set MySQL It's done , The performance is definitely not high . If there are three MySQL, a mater Write only , Two sets of salve Only responsible for reading operations , Can't performance be greatly improved ？

So master-slave replication 、 Read write separation is to support more concurrency in the database .

With the expansion of business volume 、 If it's a stand-alone deployment MySQL, It can lead to I/O The frequency is too high . Using master-slave replication 、 Read write separation can improve the availability of the database .

The principle of master-slave replication

① When Master Nodes carry out insert、update、delete In operation , It will be written to binlog in .

②salve Connect from library master Main library ,Master How many slave How many binlog dump Threads .

③ When Master Node binlog When something changes ,binlog dump The thread will notify all salve node , And the corresponding binlog Push the content to slave node .

④I/O Thread received binlog After content , Write content to local relay-log.

⑤SQL Thread reads I/O Thread writes relay-log, And according to relay-log The content of the database to do the corresponding operation .

How to realize master-slave replication

I use three virtual machines here (Linux) demonstration ,IP Namely 104(Master),106(Slave),107(Slave).

The expected effect is one master and two slaves , As shown in the figure below ：

Master To configure

Using the command line to enter mysql：

1
mysql -u root -p

Copy

Then the input root User's password ( If you forget the password, check the reset password online ~), And then create the user ：

1
//192.168.0.106 yes slave Slave IP2
GRANT REPLICATION SLAVE ON *.* to 'root'@'192.168.0.106' identified by '[email protected]';3
//192.168.0.107 yes slave Slave IP4
GRANT REPLICATION SLAVE ON *.* to 'root'@'192.168.0.107' identified by '[email protected]';5
// Refresh the configuration of the system permission table 6
FLUSH PRIVILEGES; The two users created are configuring slave You need to use .

Copy

The next step is to find mysql Configuration file for /etc/my.cnf, Add the following configuration ：

1
#  Turn on binlog2
log-bin=mysql-bin3
server-id=1044
#  Databases that need to be synchronized , If not configured, synchronize all databases 5
binlog-do-db=test_db6
# binlog The number of days the log is kept , Clear more than 10 Day's diary 7
#  Prevent log files from being too large , Cause disk space shortage 8
expire-logs-days=10

Copy

When the configuration is complete , restart mysql：

1
service mysql restart	

Copy

You can do this from the command line show master status\G; View the current binlog Log information ( Behind the useful )：

Slave To configure

Slave The configuration is relatively simple . The slave must also be a MySQL The server , So and Master equally , find /etc/my.cnf The configuration file , Add the following configuration ：

1
#  Don't talk to other people mysql service id Just repeat 2
server-id=106 Then log in to... Using the command line mysql The server ：

Copy

1
mysql -u root -p

Copy

Then type in your password and log in .

Enter into mysql after , Enter the following command again ：

1
CHANGE MASTER TO 2
MASTER_HOST='192.168.0.104',// host IP3
MASTER_USER='root',// Previously created user account 4
MASTER_PASSWORD='[email protected]',// Previously created user password 5
MASTER_LOG_FILE='mysql-bin.000001',//master The host binlog Log name 6
MASTER_LOG_POS=862,//binlog Log offset 7
master_port=3306;// port 

Copy

It's not over yet. , After setting, you need to start ：

1
#  start-up slave service 2
start slave;			

Copy

 How to check whether the startup is successful after startup ？ Use the following command ：

Copy

1
show slave status\G;	

Copy

You can see the following information （ Extract some key information ）：

Test master-slave replication

stay master Host execution sql：

1
CREATE TABLE `tb_commodity_info` (2
  `id` varchar(32) NOT NULL,3
  `commodity_name` varchar(512) DEFAULT NULL COMMENT ' Name of commodity ',4
  `commodity_price` varchar(36) DEFAULT '0' COMMENT ' commodity price ',5
  `number` int(10) DEFAULT '0' COMMENT ' The number ',6
  `description` varchar(2048) DEFAULT '' COMMENT ' Commodity Description ',7
  PRIMARY KEY (`id`)8
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT=' Commodity information table ';	

Copy

Then we can see two slave Slave synchronization also creates product information tables ：

Master slave replication is done ！

Publisher ： Full stack programmer stack length , Reprint please indicate the source ：https://javaforall.cn/111542.html Link to the original text ：https://javaforall.cn