8.12. mysqldump — A Database Backup Program

MySQL 5.0

previous page next page

8.12. mysqldump — A Database Backup Program

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 for transferring the data to another SQL server (not necessarily a MySQL server). The dump contains SQL statements to create the table or populate it, or both.

If you are doing a backup on the server, and your tables all are MyISAM tables, consider using the mysqlhotcopy instead because it can accomplish faster backups and faster restores. See Section 8.13, “mysqlhotcopy — A Database Backup Program”.

There are three general ways to invoke mysqldump:

shell> mysqldump [options] db_name [tables]
shell> mysqldump [options] --databases db_name1 [db_name2 db_name3...]
shell> mysqldump [options] --all-databases

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

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

If you run mysqldump without the --quick or --opt option, mysqldump loads the whole result set into memory before dumping the result. This can be a problem if you are dumping a big database. The --opt option is enabled by default, but can be disabled with --skip-opt.

If you are using a recent copy of the mysqldump program 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.

mysqldump supports the following options:

--help, -?

Display a help message and exit.

--add-drop-database

Add a DROP DATABASE statement before each CREATE DATABASE statement.

--add-drop-table

Add a DROP TABLE statement before each CREATE TABLE statement.

--add-locks

Surround each table dump with LOCK TABLES and UNLOCK TABLES statements. This results in faster inserts when the dump file is reloaded. See Section 7.2.16, “Speed of INSERT Statements”.

--all-databases, -A

Dump all tables in all databases. This is the same as using the --databases option and naming all the databases on the command line.

--allow-keywords

Allow creation of column names that are keywords. This works by prefixing each column name with the table name.

--character-sets-dir=path

The directory where character sets are installed. See Section 5.11.1, “The Character Set Used for Data and Sorting”.

--comments, -i

Write additional information in the dump file such as program version, server version, and host. . This option is enabled by default. To suppress additional, use --skip-comments.

--compact

Produce less verbose output. This option suppresses comments and enables the --skip-add-drop-table, --no-set-names, --skip-disable-keys, and --skip-add-locks options.

--compatible=name

Produce output that is more compatible with other database systems or with older MySQL servers. The value of name can be ansi, mysql323, mysql40, postgresql, oracle, mssql, db2, maxdb, no_key_options, no_table_options, or no_field_options. To use several values, separate them by commas. These values have the same meaning as the corresponding options for setting the server SQL mode. See Section 5.2.5, “The Server SQL Mode”.

This option does not guarantee compatibility with other servers. It only enables those SQL mode values that are currently available for making dump output more compatible. For example, --compatible=oracle does not map data types to Oracle types or use Oracle comment syntax.

--complete-insert, -c

Use complete INSERT statements that include column names.

--compress, -C

Compress all information sent between the client and the server if both support compression.

--create-options

Include all MySQL-specific table options in the CREATE TABLE statements.

--databases, -B

Dump several databases. Normally, mysqldump treats the first name argument on the command line as a database name and following names as table names. With this option, it treats all name arguments as database names. CREATE DATABASE and USE statements are included in the output before each new database.

--debug[=debug_options], -# [debug_options]

Write a debugging log. The debug_options string is often 'd:t:o,file_name'. The default is 'd:t:o,/tmp/mysqldump.trace'.

--default-character-set=charset_name

Use charset_name as the default character set. See Section 5.11.1, “The Character Set Used for Data and Sorting”. If not specified, mysqldump uses utf8.

--delayed-insert

Write INSERT DELAYED statements rather than INSERT statements.

--delete-master-logs

On a master replication server, delete the binary logs after performing the dump operation. This option automatically enables --master-data.

--disable-keys, -K

For each table, surround the INSERT statements with /*!40000 ALTER TABLEtbl_name DISABLE KEYS */; and /*!40000 ALTER TABLEtbl_name ENABLE KEYS */; statements. This makes loading the dump file faster because the indexes are created after all rows are inserted. This option is effective for MyISAM tables only.

--extended-insert, -e

Use multiple-row INSERT syntax that include several VALUES lists. This results in a smaller dump file and speeds up inserts when the file is reloaded.

--fields-terminated-by=..., --fields-enclosed-by=..., --fields-optionally-enclosed-by=..., --fields-escaped-by=..., --lines-terminated-by=...

These options are used with the -T option and have the same meaning as the corresponding clauses for LOAD DATA INFILE. See Section 13.2.5, “LOAD DATA INFILE Syntax”.

--first-slave, -x

Deprecated. Now renamed to --lock-all-tables.

--flush-logs, -F

Flush the MySQL server log files before starting the dump. This option requires the RELOAD privilege. Note that if you use this option in combination with the --all-databases (or -A) option, the logs are flushed for each database dumped. The exception is when using --lock-all-tables or --master-data: In this case, the logs are flushed only once, corresponding to the moment that all tables are locked. If you want your dump and the log flush to happen at exactly the same moment, you should use --flush-logs together with either --lock-all-tables or --master-data.

--force, -f

Continue even if an SQL error occurs during a table dump.

One use for this option is to cause mysqldump to continue executing even when it encounters a view that has become invalid because the defintion refers to a table that has been dropped. Without --force, mysqldump exits with an error message. With --force, mysqldump prints the error message, but it also writes a SQL comment containing the view definition to the dump output and continues executing.

--host=host_name, -h host_name

Dump data from the MySQL server on the given host. The default host is localhost.

--hex-blob

Dump binary columns using hexadecimal notation (for example, 'abc' becomes 0x616263). The affected data types are BINARY, VARBINARY, and BLOB. As of MySQL 5.0.13, BIT columns are affected as well.

--ignore-table=db_name.tbl_name

Do not dump the given table, which must be specified using both the database and table names. To ignore multiple tables, use this option multiple times.

--insert-ignore

Write INSERT statements with the IGNORE option.

--lock-all-tables, -x

Lock all tables across all databases. This is achieved by acquiring a global read lock for the duration of the whole dump. This option automatically turns off --single-transaction and --lock-tables.

--lock-tables, -l

Lock all tables before starting the dump. The tables are locked with READ LOCAL to allow concurrent inserts in the case of MyISAM tables. For transactional tables such as InnoDB and BDB, --single-transaction is a much better option, because it does not need to lock the tables at all.

Please note that when dumping multiple databases, --lock-tables locks tables for each database separately. So, this option does not guarantee that the tables in the dump file are logically consistent between databases. Tables in different databases may be dumped in completely different states.

--master-data[=value]

Write the binary log filename and position to the output. This option requires the RELOAD privilege and the binary log must be enabled. If the option value is equal to 1, the position and filename are written to the dump output in the form of a CHANGE MASTER statement that makes a slave server start from the correct position in the master's binary logs if you use this SQL dump of the master to set up a slave. If the option value is equal to 2, the CHANGE MASTER statement is written as an SQL comment. This is the default action if value is omitted.

The --master-data option turns on --lock-all-tables, unless --single-transaction also is specified (in which case, a global read lock is only acquired a short time at the beginning of the dump. See also the description for --single-transaction. In all cases, any action on logs happens at the exact moment of the dump. This option automatically turns off --lock-tables.

--no-autocommit

Enclose the INSERT statements for each dumped table within SET AUTOCOMMIT=0 and COMMIT statements.

--no-create-db, -n

This option suppresses the CREATE DATABASE statements that are otherwise included in the output if the --databases or --all-databases option is given.

--no-create-info, -t

Do not write CREATE TABLE statements that re-create each dumped table.

--no-data, -d

Do not write any row information for the table. This is very useful if you want to dump only the CREATE TABLE statement for the table.

--opt

This option is shorthand; it is the same as specifying --add-drop-table --add-locks --create-options --disable-keys --extended-insert --lock-tables --quick --set-charset. It should give you a fast dump operation and produce a dump file that can be reloaded into a MySQL server quickly.

The --opt option is enabled by default. To disable the options that it enables, use --skip-opt. To disable only certain of the options enabled by --opt, use their --skip forms; for example, --skip-add-drop-table or --skip-quick. Alternatively, use --skip-opt to disable the options enabled by --opt, followed by options to enable the features that you want. Options are processed in order, so the options to enable features must follow --skip-opt. For example, --skip-opt --extended-insert enables extended inserts, but --extended-insert --skip-opt does not.

--order-by-primary

Sorts each table's rows by its primary key, or its first unique index, if such an index exists. This is useful when dumping a MyISAM table to be loaded into an InnoDB table, but will make the dump itself take considerably longer.

--password[=password], -p[password]

The password to use when connecting to the server. If you use the short option form (-p), you cannot have a space between the option and the password. If you omit the password value following the --password or -p option on the command line, you are prompted for one.

Specifying a password on the command line should be considered insecure. See Section 5.9.6, “Keeping Your Password Secure”.

--port=port_num, -P port_num

The TCP/IP port number to use for the connection.

--protocol={TCP|SOCKET|PIPE|MEMORY}

The connection protocol to use.

--quick, -q

This option is useful for dumping large tables. It forces mysqldump to retrieve rows for a table from the server a row at a time rather than retrieving the entire row set and buffering it in memory before writing it out.

--quote-names, -Q

Quote database, table, and column names within ‘`’ characters. If the ANSI_QUOTES SQL mode is enabled, names are quoted within ‘"’ characters. This option is enabled by default. It can be disabled with --skip-quote-names, but this option should be given after any option such as --compatible that may enable --quote-names.

--result-file=file, -r file

Direct output to a given file. This option should be used on Windows to prevent newline ‘\n’ characters from being converted to ‘\r\n’ carriage return/newline sequences. The result file is created and its contents overwritten, even if an error occurs while generating the dump. The previous contents are lost.

--routines, -R

Dump stored routines (functions and procedures) from the dumped databases. The output generated by using --routines contains CREATE PROCEDURE and CREATE FUNCTION statements to re-create the routines. However, these statements do not include attributes such as the routine creation and modification timestamps. This means that when the routines are reloaded, they will be created with the timestamps equal to the reload time.

If you require routines to be re-created with their original timestamp attributes, do not use --routines. Instead, dump and reload the contents of the mysql.proc table directly, using a MySQL account that has appropriate privileges for the mysql database.

This option was added in MySQL 5.0.13. Before that, stored routines are not dumped. Routine DEFINER values are not dumped until MySQL 5.0.20. This means that before 5.0.20, when routines are reloaded, they will be created with the definer set to the reloading user. If you require routines to be re-created with their original definer, dump and load the contents of the mysql.proc table directly as described earlier.

--set-charset

Add SET NAMESdefault_character_set to the output. This option is enabled by default. To suppress the SET NAMES statement, use --skip-set-charset.

--single-transaction

This option issues a BEGIN SQL statement before dumping data from the server. It is useful only with transactional tables such as InnoDB and BDB, because then it dumps the consistent state of the database at the time when BEGIN was issued without blocking any applications.

When using this option, you should keep in mind that only InnoDB tables are dumped in a consistent state. For example, any MyISAM or MEMORY tables dumped while using this option may still change state.

The --single-transaction option and the --lock-tables option are mutually exclusive, because LOCK TABLES causes any pending transactions to be committed implicitly.

This option is not supported for MySQL Cluster tables; the results cannot be guaranteed to be consistent due to the fact that the NDBCluster storage engine supports only the READ_COMMITTED transaction isolation level. You should always use NDB backup and restore instead.

To dump big tables, you should combine this option with --quick.

--skip-opt

See the description for the --opt option.

--socket=path, -S path

For connections to localhost, the Unix socket file to use, or, on Windows, the name of the named pipe to use.

--skip-comments

See the description for the --comments option.

--ssl*

Options that begin with --ssl specify whether to connect to the server via SSL and indicate where to find SSL keys and certificates. See Section 5.9.7.3, “SSL Command Options”.

--tab=path, -T path

Produce tab-separated data files. For each dumped table, mysqldump creates a tbl_name.sql file that contains the CREATE TABLE statement that creates the table, and a tbl_name.txt file that contains its data. The option value is the directory in which to write the files.

By default, the .txt data files are formatted using tab characters between column values and a newline at the end of each line. The format can be specified explicitly using the --fields-xxx and --lines--xxx options.

Note: This option should be used only when mysqldump is run on the same machine as the mysqld server. You must have the FILE privilege, and the server must have permission to write files in the directory that you specify.

--tables

Override the --databases or -B option. All name arguments following the option are regarded as table names.

--triggers

Dump triggers for each dumped table. This option is enabled by default; disable it with --skip-triggers. This option was added in MySQL 5.0.11. Before that, triggers are not dumped.

--tz-utc

Add SET TIME_ZONE='+00:00' to the dump file so that TIMESTAMP columns can be dumped and reloaded between servers in different time zones. Without this option, TIMESTAMP columns are dumped and reloaded in the time zones local to the source and destination servers, which can cause the values to change. --tz-utc also protects against changes due to daylight saving time. --tz-utc is enabled by default. To disable it, use --skip-tz-utc. This option was added in MySQL 5.0.15.

--user=user_name, -u user_name

The MySQL username to use when connecting to the server.

--verbose, -v

Verbose mode. Print more information about what the program does.

--version, -V

Display version information and exit.

--where='where_condition', -w 'where_condition'

Dump only rows selected by the given WHERE condition. Note that quotes around the condition are mandatory if it contains spaces or other characters that are special to your command interpreter.

Examples:

--where="user='jimf'" -w"userid>1" -w"userid<1"

--xml, -X

Write dump output as well-formed XML.

You can also set the following variables by using --var_name=value syntax:

max_allowed_packet

The maximum size of the buffer for client/server communication. The maximum is 1GB.

net_buffer_length

The initial size of the buffer for client/server communication. When creating multiple-row-insert statements (as with option --extended-insert or --opt), mysqldump creates rows up to net_buffer_length length. If you increase this variable, you should also ensure that the net_buffer_length variable in the MySQL server is at least this large.

It is also possible to set variables by using --set-variable=var_name=value or -Ovar_name=value syntax. This syntax is deprecated.

The most common use of mysqldump is probably for making a backup of an entire database:

shell> mysqldump --opt db_name > backup-file.sql

You can read 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 MySQL 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 just needs to acquire 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 and only if one long updating statement is running when the FLUSH statement is issued, the MySQL server may get stalled until that long statement finishes, and then the dump becomes lock-free. If the update statements that the MySQL 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 (see Section 5.12.3, “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 simultaneous use of --master-data and --single-transaction provides a convenient way to make an online backup suitable for point-in-time recovery if tables are stored in the InnoDB storage engine.

For more information on making backups, see Section 5.10.1, “Database Backups”, and Section 5.10.2, “Example Backup and Recovery Strategy”.

previous page start next page