How to back up MySQL_ The most complete MySQL backup method in history

Hello everyone , I meet you again , I'm your friend, Quan Jun .

I have

The backup methods used are :mysqldump、mysqlhotcopy、BACKUP TABLE 、SELECT INTO

OUTFILE, Or back up binary logs (binlog), You can also copy data files and related configuration files directly .MyISAM

Tables are saved as files , Therefore, it is relatively easy to back up , Several methods mentioned above can be used .Innodb All tables are stored in the same data file ibdata1

in ( It could be multiple files , Or a separate tablespace file ), Relatively bad backup , The free solution can be to copy data files 、 Backup binlog, Or use

mysqldump.

1.mysqldump Backup

mysqldump Is to use SQL Level backup mechanism , It leads the data table to SQL Script files , In different MySQL The upgrade between versions is relatively appropriate , This is also the most common backup method .

Example :mysqldump -uroot -p database table > /home/jobs/back.sql

mysqldump It can also be used for incremental backup ,mysqldump There are many related parameters online , Let's not repeat them here

2.mysqlhotcopy Backup

mysqlhotcopy It's a PERL Program . It USES LOCK TABLES、FLUSH

TABLES and cp or scp

To quickly back up the database . It's the fastest way to back up a database or a single table , But it can only run in database files ( Include data table definition file 、 Data files 、 Index file ) On the same machine .

mysqlhotcopy Can only be used for backup MyISAM, And it can only run in class Unix and NetWare On the system .

mysqlhotcopy It supports copying multiple databases at one time , It also supports regular expressions .

Example ： root#/usr/local/mysql/bin/mysqlhotcopy -h=localhost -u=root

-p=123456 database /tmp ( Put the database directory database copy to /tmp

Next )root#/usr/local/mysql/bin/mysqlhotcopy -h=localhost -u=root -p=123456

db_name_1 … db_name_n /tmproot#/usr/local/mysql/bin/mysqlhotcopy

-h=localhost -u=root -p=123456 db_name./regex/

/tmp Please refer to the manual for more detailed usage , Or call the following command to view mysqlhotcopy With the help of the ：

perldoc /usr/local/mysql/bin/mysqlhotcopy Be careful , Want to use mysqlhotcopy, There must be

SELECT、RELOAD( To execute FLUSH TABLES) jurisdiction , And must be able to read datadir/db_name Directory permissions .

Restore mysqlhotcopy The whole database directory is backed up , It can be directly copied to mysqld

designated datadir ( Here it is. /usr/local/mysql/data/) Under the directory , At the same time, attention should be paid to the issue of permissions , Here's an example ： root#cp

-rf db_name /usr/local/mysql/data/root#chown -R nobody:nobody

/usr/local/mysql/data/ ( take db_name The owner of the directory is changed to mysqld Run the user )

3.SQL Syntax backup

3.1 Backup BACKUP TABLE Grammar is actually the same as mysqlhotcopy

It works pretty much the same way , All locked watches , Then copy the data file . It enables online backup , But the effect is not ideal , Therefore, it is not recommended . It only copies table structure files and data files , Do not copy the citation at the same time

Pieces of , Therefore, the recovery time is relatively slow . Example ： BACK TABLE tbl_name TO ‘/tmp/db_name/’; Be careful , There must be FILE

Permission to execute this SQL, And the directory /tmp/db_name/ Must be able to be mysqld Users can write , The exported file cannot overwrite the existing file , To avoid safety issues .

Recovery BACKUP TABLE Method to back up the file , Can run RESTORE TABLE Statement to restore the data table . Example ： RESTORE TABLE FROM ‘/tmp/db_name/’; Permission requirements are similar to those described above .

3.2 SELECT INTO OUTFILE Is to export the data into a common text file , You can customize how fields are spaced , It is convenient to process these data . Example ：

SELECT INTO OUTFILE ‘/tmp/db_name/tbl_name.txt’ FROM tbl_name; Be careful , There must be

FILE Permission to execute this SQL, And the documents /tmp/db_name/tbl_name.txt Must be able to be mysqld

Users can write , The exported file cannot overwrite the existing file , To avoid safety issues .

use SELECT INTO OUTFILE Method to back up the file , Can run LOAD DATA INFILE Statement to restore the data table . Example ： LOAD

DATA INFILE ‘/tmp/db_name/tbl_name.txt’ INTO TABLE

tbl_name; Permission requirements are similar to those described above . Before pouring data , The data table must already exist . If you worry about data duplication , Can increase REPLACE

Keyword to replace existing records or use IGNORE Keywords to ignore them .

4. Enable binary logging (binlog)

use binlog The method is relatively more flexible , Save the mind and save labour , It can also support incremental backup .

Enable binlog It must be restarted mysqld. First , close mysqld, open my.cnf, Add the following lines ：

server-id = 1

log-bin = binlog

log-bin-index = binlog.index

Then start mysqld That's all right. . During operation, it will produce binlog.000001 as well as binlog.index, The previous file is mysqld

Record all updates to the data , The following file is all binlog The index of , Can not be easily deleted . About binlog Please refer to the manual for more information .

When backup is required , You can do it first SQL sentence , Give Way mysqld Terminate the current binlog

Writing , You can back up the files directly , In this way, the purpose of incremental backup can be achieved ： FLUSH LOGS; If it is a slave server in the backup replication system , You should also back up

master.info and relay-log.info file .

Back up binlog Documents available MySQL Tools provided mysqlbinlog Check it out. , Such as ：

/usr/local/mysql/bin/mysqlbinlog /tmp/binlog.000001 This tool allows you to display all the data under the specified database

SQL sentence , And it can also limit the time range , Quite handy , Please refer to the manual for details .

When you recover , You can do this with statements like the following ： /usr/local/mysql/bin/mysqlbinlog /tmp/binlog.000001

| mysql -uyejr -pyejr db_name hold mysqlbinlog Output SQL Statement executes it directly as input .

If you have a free machine , This is a good way to back up . As a result of slave The performance requirements of the machine are not so high , So the cost is low , Incremental backup can be realized at a low cost, and some data query pressure can be shared , Why not do it ？

5. Copy files

Compared with the previous methods, the direct backup of data files , Backing up data files is the most direct 、 Fast 、 convenient , The disadvantage is that incremental backup is basically impossible .

To ensure data consistency , Need to be in front of the backrest file , Perform the following SQL sentence ： FLUSH TABLES WITH READ

LOCK; That is to refresh the data in the memory to the disk , Lock the data table at the same time , To ensure that no new data is written during the copy process . The recovery of data backed up by this method is also very simple , Copy directly back to

Under the original database directory .

Be careful , about Innodb Type table , You also need to back up its log files , namely ib_logfile* file . Because when Innodb When the watch is damaged , You can rely on these log files to recover .

6. utilize rsync Backup

rsync As a synchronization tool, it can also be used for backup , But you need to configure the server side and the client side

Example rsync -vzrtopg –progress –delete [email protected]::root /tmp/

The disadvantage is that rsync It is an incremental backup based on the file modification time , Therefore, all backup databases are fully backed up , And the configuration is troublesome .

Publisher ： Full stack programmer stack length , Reprint please indicate the source ：https://javaforall.cn/150622.html Link to the original text ：https://javaforall.cn