mysqldump

The mysqldump client is a backup program originally written by Igor Romanenko. It can be used to dump a database or a collection of databases for backup or transfer to another database server (not necessarily MariaDB or MySQL). The dump typically contains SQL statements to create the table, populate it, or both. However, mysqldump can also be used to generate files in CSV, other delimited text, or XML format.

If you are doing a backup on the server and your tables all are MyISAM tables, consider using mysqlhotcopy instead because it can accomplish faster backups and faster restores.

mysqldump dumps triggers along with tables, as these are part of the table definition. However, stored procedures, views, and events are not, and need extra parameters to be recreated explicitly (for example, --routines and --events). Procedures and functions are however also part of the system tables (for example mysql.proc).

mysqldump supports the enhancements for START TRANSACTION WITH CONSISTENT SNAPSHOT.

Performance

mysqldump doesn't usually consume much CPU resources on modern hardware as by default it uses a single thread. This method is good for a heavily loaded server.

Disk input/outputs per second (IOPS), can however increase for multiple reasons. When you back-up on the same device as the database, this produces unnecessary random IOPS. The dump is done sequentially, on a per table basis, causing a full-table scan and many buffer page misses on tables that are not fully cached in memory.

It's recommended that you back-up from a network location to remove disk IOPS on the database server, but it is vital to use a separate network card to keep network bandwidth available for regular traffic.

Although mysqldump will by default preserve your resources for regular spindle disks and low-core hardware, this doesn't mean that concurrent dumps cannot benefit from hardware architecture like SAN, flash storage, low write workload. The back-up time would benefit from a tool such as MyDumper.

Usage

There are four general ways to invoke mysqldump:

shell> mysqldump [options] db_name [tbl_name ...]
shell> mysqldump [options] --databases db_name ...
shell> mysqldump [options] --all-databases
shell> mysqldump [options] --system=[option_list]

If you do not name any tables following db_name or if you use the --databases or --all-databases option, entire databases are dumped.

mysqldump does not dump the INFORMATION_SCHEMA (or PERFORMANCE_SCHEMA, if enabled) database by default. MariaDB dumps the INFORMATION_SCHEMA if you name it explicitly on the command line, although currently you must also use the --skip-lock-tables option.

To see a list of the options your version of mysqldump supports, execute mysqldump --help.

Row by Row vs. Buffering

mysqldump can retrieve and dump table contents row by row, or it can retrieve the entire content from a table and buffer it in memory before dumping it. Buffering in memory can be a problem if you are dumping large tables. To dump tables row by row, use the --quick option (or --opt, which enables --quick). The --opt option (and hence --quick) is enabled by default, so to enable memory buffering, use --skip-quick.

mysqldump in MariaDB 10.3 and Higher

mysqldump in MariaDB 10.3 includes logic to cater for the mysql.transaction_registry table. mysqldump from an earlier MariaDB release cannot be used on MariaDB 10.3 and beyond.

mysqldump and Old Versions of MySQL

If you are using a recent version of mysqldump to generate a dump to be reloaded into a very old MySQL server, you should not use the --opt or --extended-insert option. Use --skip-opt instead.

Options

mysqldump supports the following options:

-- Dump completed on DATE

mysql> STOP SLAVE SQL_THREAD;

mysql> SHOW SLAVE STATUS;

shell> mysqldump --master-data=2 --all-databases > dumpfile

mysql> START SLAVE;

shell> mysql < dumpfile

mysql> CHANGE MASTER TO MASTER_LOG_FILE = ´file_name´, MASTER_LOG_POS = file_pos;

--where="user=´jimf´"

-w"userid>1"

-w"userid<1"

Group Options

Some mysqldump options are shorthand for groups of other options:

• Use of --opt is the same as specifying --add-drop-table, --add-locks, --create-options, --disable-keys, --extended-insert, --lock-tables, --quick, and --set-charset. All of the options that --opt stands for also are on by default because --opt is on by default.
• Use of --compact is the same as specifying --skip-add-drop-table, --skip-add-locks, --skip-comments, --skip-disable-keys, and --skip-set-charset options.

To reverse the effect of a group option, uses its --skip-xxx form (--skip-opt or --skip-compact). It is also possible to select only part of the effect of a group option by following it with options that enable or disable specific features. Here are some examples:

• To select the effect of --opt except for some features, use the --skip option for each feature. To disable extended inserts and memory buffering, use --opt --skip-extended-insert --skip-quick. (Actually, --skip-extended-insert --skip-quick is sufficient because --opt is on by default.)
• To reverse --opt for all features except index disabling and table locking, use --skip-opt --disable-keys --lock-tables.

When you selectively enable or disable the effect of a group option, order is important because options are processed first to last. For example, --disable-keys --lock-tables --skip-opt would not have the intended effect; it is the same as --skip-opt by itself.

Special Characters in Option Values

Some options, like --lines-terminated-by, accept a string. The string can be quoted, if necessary. For example, on Unix systems this is the option to enclose fields within double quotes:

--fields-enclosed-by='"'

An alternative to specify the hexadecimal value of a character. For example, the following syntax works on any platform:

--fields-enclosed-by=0x22

Option Files

In addition to reading options from the command-line, mysqldump can also read options from option files. If an unknown option is provided to mysqldump in an option file, then it is ignored.

The following options relate to how MariaDB command-line tools handles option files. They must be given as the first argument on the command-line:

In MariaDB 10.2 and later, mysqldump is linked with MariaDB Connector/C. However, MariaDB Connector/C does not yet handle the parsing of option files for this client. That is still performed by the server option file parsing code. See MDEV-19035 for more information.

Option Groups

mysqldump reads options from the following option groups from option files:

NULL, ´NULL´, and Empty Values in XML

For a column named column_name, the NULL value, an empty string, and the string value ´NULL´ are distinguished from one another in the output generated by this option as follows.

The output from the mysql client when run using the --xml option also follows the preceding rules.

XML output from mysqldump includes the XML namespace, as shown here :

shell> mysqldump --xml -u root world City
<?xml version="1.0"?>
<mysqldump xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<database name="world">
<table_structure name="City">
<field Field="ID" Type="int(11)" Null="NO" Key="PRI" Extra="auto_increment" />
<field Field="Name" Type="char(35)" Null="NO" Key="" Default="" Extra="" />
<field Field="CountryCode" Type="char(3)" Null="NO" Key="" Default="" Extra="" />
<field Field="District" Type="char(20)" Null="NO" Key="" Default="" Extra="" />
<field Field="Population" Type="int(11)" Null="NO" Key="" Default="0" Extra="" />
<key Table="City" Non_unique="0" Key_name="PRIMARY" Seq_in_index="1" Column_name="ID"
Collation="A" Cardinality="4079" Null="" Index_type="BTREE" Comment="" />
<options Name="City" Engine="MyISAM" Version="10" Row_format="Fixed" Rows="4079"
Avg_row_length="67" Data_length="273293" Max_data_length="18858823439613951"
Index_length="43008" Data_free="0" Auto_increment="4080"
Create_time="2007-03-31 01:47:01" Update_time="2007-03-31 01:47:02"
Collation="latin1_swedish_ci" Create_options="" Comment="" />
</table_structure>
<table_data name="City">
<row>
<field name="ID">1</field>
<field name="Name">Kabul</field>
<field name="CountryCode">AFG</field>
<field name="District">Kabol</field>
<field name="Population">1780000</field>
</row>
...
<row>
<field name="ID">4079</field>
<field name="Name">Rafah</field>
<field name="CountryCode">PSE</field>
<field name="District">Rafah</field>
<field name="Population">92020</field>
</row>
</table_data>
</database>
</mysqldump>

Restoring

To restore a backup created with mysqldump, use the mysql client to import the dump, for example:

mysql db_name < backup-file.sql

Variables

You can also set the following variables (--variable-name=value) and boolean options {FALSE|TRUE} by using:

Examples

A common use of mysqldump is for making a backup of an entire database:

shell> mysqldump db_name > backup-file.sql

You can load the dump file back into the server like this:

shell> mysql db_name < backup-file.sql

Or like this:

shell> mysql -e "source /path-to-backup/backup-file.sql" db_name

mysqldump is also very useful for populating databases by copying data from one MariaDB server to another:

shell> mysqldump --opt db_name | mysql --host=remote_host -C db_name

It is possible to dump several databases with one command:

shell> mysqldump --databases db_name1 [db_name2 ...] > my_databases.sql

To dump all databases, use the --all-databases option:

shell> mysqldump --all-databases > all_databases.sql

For InnoDB tables, mysqldump provides a way of making an online backup:

shell> mysqldump --all-databases --single-transaction all_databases.sql

This backup acquires a global read lock on all tables (using FLUSH TABLES WITH READ LOCK) at the beginning of the dump. As soon as this lock has been acquired, the binary log coordinates are read and the lock is released. If long updating statements are running when the FLUSH statement is issued, the MariaDB server may get stalled until those statements finish. After that, the dump becomes lock free and does not disturb reads and writes on the tables. If the update statements that the MariaDB server receives are short (in terms of execution time), the initial lock period should not be noticeable, even with many updates.

For point-in-time recovery (also known as “roll-forward,” when you need to restore an old backup and replay the changes that happened since that backup), it is often useful to rotate the binary log or at least know the binary log coordinates to which the dump corresponds:

shell> mysqldump --all-databases --master-data=2 > all_databases.sql

Or:

shell> mysqldump --all-databases --flush-logs --master-data=2 > all_databases.sql

The --master-data and --single-transaction options can be used simultaneously, which provides a convenient way to make an online backup suitable for use prior to point-in-time recovery if tables are stored using the InnoDB storage engine.

See Also

• Mariabackup
• MariaDB Enterprise Backup
• Upgrading to a newer major version of MariaDB (video)