Migrate large databases to Azure Database for MySQL using mydumper/myloader

APPLIES TO: Azure Database for MySQL - Single Server Azure Database for MySQL - Flexible Server

Azure Database for MySQL is a managed service that you use to run, manage, and scale highly available MySQL databases in the cloud. To migrate MySQL databases larger than 1 TB to Azure Database for MySQL, consider using community tools such as mydumper/myloader, which provide the following benefits:

  • Parallelism, to help reduce the migration time.
  • Better performance, by avoiding expensive character set conversion routines.
  • An output format, with separate files for tables, metadata etc., that makes it easy to view/parse data. Consistency, by maintaining snapshot across all threads.
  • Accurate primary and replica log positions.
  • Easy management, as they support Perl Compatible Regular Expressions (PCRE) for specifying database and tables inclusions and exclusions.
  • Schema and data goes together. Don't need to handle it separately like other logical migration tools.

This quickstart shows you how to install, back up, and restore a MySQL database by using mydumper/myloader.

Prerequisites

Before you begin migrating your MySQL database, you need to:

  1. Create an Azure Database for MySQL server by using the Azure portal.

  2. Create an Azure VM running Linux by using the Azure portal (preferably Ubuntu).

    Note

    Prior to installing the tools, consider the following points:

    • If your source is on-premises and has a high bandwidth connection to Azure (using ExpressRoute), consider installing the tool on an Azure VM.
    • If you have a challenge in the bandwidth between the source and target, consider installing mydumper near the source and myloader near the target server. You can use tools Azcopy to move the data from on-premises or other cloud solutions to Azure.
  3. Install mysql client, do the following steps:

    • Update the package index on the Azure VM running Linux by running the following command:
      $ sudo apt update
      
    • Install the mysql client package by running the following command:
      $ sudo apt install mysql-client
      

Install mydumper/myloader

To install mydumper/myloader, do the following steps.

  1. Depending on your OS distribution, download the appropriate package for mydumper/myloader, running the following command:

  2. $ wget https://github.com/maxbube/mydumper/releases/download/v0.10.1/mydumper_0.10.1-2.$(lsb_release -cs)_amd64.deb
    

    Note

    $(lsb_release -cs) helps to identify your distribution.

  3. To install the .deb package for mydumper, run the following command:

    $ dpkg -i mydumper_0.10.1-2.$(lsb_release -cs)_amd64.deb
    

    Tip

    The command you use to install the package will differ based on the Linux distribution you have as the installers are different. The mydumper/myloader is available for following distributions Fedora, RedHat , Ubuntu, Debian, CentOS , openSUSE and MacOSX. For more information, see How to install mydumper

Create a backup using mydumper

  • To create a backup using mydumper, run the following command:

    $ mydumper --host=<servername> --user=<username> --password=<Password> --outputdir=./backup --rows=100000 --compress --build-empty-files --threads=16 --compress-protocol --trx-consistency-only --ssl  --regex '^(<Db_name>\.)' -L mydumper-logs.txt
    

This command uses the following variables:

  • --host: The host to connect to

  • --user: Username with the necessary privileges

  • --password: User password

  • --rows: Try to split tables into chunks of this many rows

  • --outputdir: Directory to dump output files to

  • --regex: Regular expression for Database matching.

  • --trx-consistency-only: Transactional consistency only

  • --threads: Number of threads to use, default 4. Recommended a use a value equal to 2x of the vCore of the computer.

    Note

    For more information on other options, you can use with mydumper, run the following command: mydumper --help . For more details see, mydumper\myloader documentation
    To dump multiple databases in parallel, you can modiffy regex variable as shown in the example: regex ’^(DbName1.|DbName2.)

Restore your database using myloader

  • To restore the database that you backed up using mydumper, run the following command:

    $ myloader --host=<servername> --user=<username> --password=<Password> --directory=./backup --queries-per-transaction=500 --threads=16 --compress-protocol --ssl --verbose=3 -e 2>myloader-logs.txt
    

This command uses the following variables:

  • --host: The host to connect to
  • --user: Username with the necessary privileges
  • --password: User password
  • --directory: Location where the backup is stored.
  • --queries-per-transaction: Recommend setting to value not more than 500
  • --threads: Number of threads to use, default 4. Recommended a use a value equal to 2x of the vCore of the computer

Tip

For more information on other options you can use with myloader, run the following command: myloader --help

After the database is restored, it’s always recommended to validate the data consistency between the source and the target databases.

Note

Submit any issues or feedback regarding the mydumper/myloader tools here.

Next steps