Use a custom import to set up replication from large external databases

MySQL   |  PostgreSQL   |  SQL Server

This page describes the process for setting up external server replication using a custom import. These steps are the best option when you need to replicate from a large external database.

You must complete all the steps on this page. When finished, you can administer and monitor the replica the same way as you would any other Cloud SQL instance.

This process is supported only for external servers that are configured to use global transaction identifier (GTID)-based replication. Before replication can be initiated, you need to load data from the external server into the Cloud SQL replica. If you don't use GTID-based replication, then Cloud SQL can't identify the exact binary log position from which to begin replication. If you can't use GITD-based replication, then you need to configure your dump tool to institute a global read-only lock during the dump process.

GTID replication isn't supported for certain versions of Amazon RDS, Amazon Aurora 5.6, and early releases of Amazon Aurora 5.7.

Before you begin

Before you start, you should have configured the external server, created the source representation instance, and set up the Cloud SQL replica.

Update permissions for the replication user

The replication user on the external server is configured to accept connections from any host (%). You should update this user account so that it can only be used with the Cloud SQL replica. Open a terminal on the source database server and enter these commands:

mysql Client

    UPDATE mysql.user
    SET Host='NEW_HOST' WHERE Host='OLD_HOST' AND User='USERNAME';
    GRANT REPLICATION SLAVE, EXECUTE ON *.*
    TO 'GCP_USERNAME'@'HOST';
    FLUSH PRIVILEGES;

example

UPDATE mysql.user
SET Host='192.0.2.0' WHERE Host='%' AND User='replicationUser';
GRANT REPLICATION SLAVE, EXECUTE ON *.*
TO 'gcp_user'@'gmail.com';
FLUSH PRIVILEGES;

Property	Description
NEW_HOST	Specify the outgoing IP of the Cloud SQL replica.
OLD_HOST	The current value assigned to Host that you want to change.
USERNAME	The replication user account on the external server.
GCP_USERNAME	The username for the GCP user account.
HOST	The hostname for the GCP user account.

Set up the Cloud SQL replica as a primary instance

Because Cloud SQL replica instances are read-only, in order to perform a custom import, you need to promote the Cloud SQL replica to a standalone instance. Once the initial data import is complete, you demote the instance back to a replica.

Perform a custom dump and import

In this section, we show you how to create the dump file and import it into the eventual Cloud SQL replica using mydumper or the mysqldump client utilities.

When you dump the data, you might need to exclude MySQL generic databases, including mysql, and sys, if they exist on the source instance. Otherwise, the data import fails. See How to exclude (or include) databases?.

Use mydumper and myloader

To create a dump file and import it to Cloud SQL:

1. Create a dump file of the external server database using mydumper.

      $ mydumper -u USERNAME -p PASSWORD \
                 --threads=16 -o ./backup \
                 -h HOST \
                 --no-locks \
                 --regex '^(?!(mysql\.|sys\.))'
   

   Property	Description
   USERNAME	The name of the replication user account or user account on the external server that has database read permissions.
   PASSWORD	Replication user password.
   HOST	The IPv4 or DNS address for the external server.

2. Import the data into the Cloud SQL instance using myloader.

    $ myloader -u REPLICA_USERNAME -p REPLICA_PASSWORD \
               --threads=16 \
               -d ./backup -h HOST -o
   

   Property	Description
   REPLICA_USERNAME	The user account on the Cloud SQL instance.
   REPLICA_PASSWORD	Cloud SQL instance user password.
   HOST	The IPv4 for the Cloud SQL instance.

3. Write down the GTID or binlog information of the data dump. You need this information when configuring the replication with the stored procedures.

   To get the GTID or binlog information of the data dump, run the following command:

     sudo cat ./backup/metadata
   

Use mysqldump

1. Create a dump using mysqldump:

   mysqldump

   mysqldump \
       --host=EXTERNAL_HOST \
       --port=EXTERNAL_PORT \
       --user=USERNAME\
       --password=PASSWORD \
       --databases=DATABASE_LIST  \
       --hex-blob \
       --master-data=EXTERNAL_DATA  \
       --no-autocommit \
       --default-character-set=utf8mb4 \
       --single-transaction \
       GTID_PURGED \
       ADD_DROP_TABLE \
       ROUTINES \
       COMPRESS \
       GZIP
   

   Property	Description
   EXTERNAL_HOST	The IPv4 or DNS address for the external server.
   EXTERNAL_PORT	The port for the external server. If the external server is hosted on Cloud SQL, this is 3306.
   USERNAME	The name of the replication user account or user account on the external server that has database read permissions.
   USER_PASSWORD	Replication user password.
   DATABASE_LIST	Space-separated list of all databases on the external server, except for the system databases (sys,mysql, performance_schema, and information_schema). Use the SHOW DATABASES MySQL command to list your databases.
   EXTERNAL_DATA	If your external server does not support GTID, and you have permission to access the global read lock on it, use --master-data=1. Otherwise, don't use this property.
   GTID_PURGED	If your external server supports GTID, use --set-gtid-purged=on; otherwise, don't use this property.
   ADD_DROP_TABLE	If you want to add a DROP TABLE statement before each CREATE TABLE statement, include --add-drop-table.
   ROUTINES	If you want to show stored routines, such as procedures and functions, in the output for dumped databases, include --routines.
   COMPRESS	If you want to compress all information sent between the Cloud SQL replica and the external server, use --compress.
   GZIP	If you want to compress the dump file even more, use | gzip. If your database contains data that does not compress well, such as binary incompressible data or JPG images, don't use this.

   example

   
   mysqldump \
       --host=192.0.2.1 \
       --port=3306 \
       --user=replicationUser \
       --password \
       --databases guestbook journal \
       --hex-blob \
       --master-data=1 \
       --no-autocommit \
       --default-character-set=utf8mb4 \
       --single-transaction \
       --compress \
       | gzip
   

2. Write down the GTID or binlog information of the data dump. You need this information to configure the replication with the Cloud SQL stored procedures.

   For the GTID, look for a line similar to the following:

      SET @@GLOBAL.GTID_PURGED='32eb1e6a-17b6-11ea-a39e-06d94ac3ec98:1-33496';
   

   For the binlog, look for a line similar to the following:

      CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin-changelog.033877', MASTER_LOG_POS=360;
   

3. Remove the following lines in the dump file that require super privileges. Since Cloud SQL users don't have super privileges, these lines cause the import to fail.

   For GTID-based replication: Remove the SET GTID_PURGED statement along with the session variable setting statement in the dump. For example:

      ...
      SET @MYSQLDUMP_TEMP_LOG_BIN = @@SESSION.SQL_LOG_BIN;
      SET @@SESSION.SQL_LOG_BIN= 0;
      ...
      SET @@GLOBAL.GTID_PURGED='32eb1e6a-17b6-11ea-a39e-06d94ac3ec98:1-33496';
      ...
      SET @@SESSION.SQL_LOG_BIN=@MYSQLDUMP_TEMP_LOG_BIN;
   

   For binlog-based replication, remove the CHANGE MASTER statement. For example:

      ...
      CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin-changelog.033877', MASTER_LOG_POS=360;
       ...
   

4. Import the data into the Cloud SQL replica, using the mysql CLI:

   mysql

   mysql -h REPLICA_HOST -u REPLICA_USER \
   -p REPLICA_DATABASE_NAME RESULT_FILE
   

   Property	Description
   REPLICA_HOST	Host on which MySQL server is located.
   REPLICA_USER	MySQL user name to use when connecting to the server.
   REPLICA_DATABASE_NAME	Name of the database where the data is located.
   RESULT_FILE	Name of the dump file to import.

   example

     mysql -h 255.255.255.255 -u replica_username -p replica_db < result.sql
   

You can also import the dump file using a Google Cloud bucket. See Importing data from a SQL dump file into Cloud SQL.

Demote the Cloud SQL instance

To demote the Cloud SQL instance to a Cloud SQL replica, use the demoteMaster method on the instance.

1. Prepare a request JSON file with the name of the instance you want to demote.

   Source JSON

    {
       "demoteMasterContext": {
         "masterInstanceName": SOURCE_REPRESENTATION_INSTANCE_NAME,
         "skipReplicationSetup": true
         }
    }
   

   Property	Description
   SOURCE_REPRESENTATION_INSTANCE_NAME	The name of the source representation instance.

   example

      {
        "demoteMasterContext": {
          "masterInstanceName": "cloudsql-source-instance",
          "skipReplicationSetup": true
        }
      }
   

2. Open a terminal and use the following commands to invoke demoteMaster:

   curl

     gcloud auth login
     ACCESS_TOKEN="$(gcloud auth print-access-token)"
     curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
       --header 'Content-Type: application/json' \
       --data @JSON_PATH \
       -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/PROJECT-ID/instances/INSTANCE-NAME/demoteMaster
   

   Property	Description
   JSON_PATH	The path to the JSON file.
   PROJECT_ID	The ID of your project in Google Cloud.
   INSTANCE-NAME	The name of the instance to demote.

   example

      gcloud auth login
      ACCESS_TOKEN="$(gcloud auth print-access-token)"
      curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
        --header 'Content-Type: application/json' \
        --data @./source.json \
        -X POST \
      https://sqladmin.googleapis.com/sql/v1beta4/projects/MyProject/instances/cloudsql-replica-instance/demoteMaster
   

What you should see when you finish

To ensure your instances were set up correctly, go to the Cloud SQL Instances page.

You should see your source representation instance and Cloud SQL replica. They look similar to the following:

Instance ID	Type	Public IP
(-) source-representation-instance	MySQL external primary	10.68.48.3:3306
replica-instance	MySQL read replica	34.66.48.59

Start replication on the Cloud SQL instance

Note: IP addresses should be preserved in the Promote or Demote steps. If another operation comes in between the steps, then the IP address of the Cloud SQL instance may change. If replication is broken because of a connectivity issue, check the outgoing IP addresses and make sure that the appropriate IP addresses are allowlisted on the external source.

This step uses Cloud SQL stored procedures. The Cloud SQL stored procedures are installed after calling the demoteMaster request. They are removed after calling promoteReplica. For more information, see Stored procedures for replication management.

1. Log on to the replica instance. For more information, see Connecting using a database client from a local machine.

2. Use the mysql.resetMaster stored procedure to reset replication settings.

    mysql> call mysql.resetMaster();
   

3. Configure the replication. This step requires the GTID or binlog information that you previously wrote down.

   GTID

   1. Configure the gtid_purged field with the mysql.skipTransactionWithGtid(GTID_TO_SKIP) stored procedure.

   Property	Description
   GTID_TO_SKIP	The GTID set value to configure.

   For example:

       mysql> call mysql.skipTransactionWithGtid('32eb1e6a-17b6-11ea-a39e-06d94ac3ec98:1-33496');
   

   1. Run the mysql.setupExternalSourceAutoPosition(HOST, PORT, USER_NAME, USER_PASSWORD, MASTER_AUTO_POSITION, USE_SSL, USE_SSL_CLIENT_AUTH) stored procedure.

   Property	Description
   HOST	Source endpoint.
   PORT	Source port.
   USER_NAME	Source user.
   USER_PASSWORD	Source user password.
   MASTER_AUTO_POSITION	Value of the master_auto_position parameter. Possible values are 0, 1.
   USE_SSL	Whether to use SSL-based replication. Possible values are true, false. If true, you need to set the caCertificate field in the DemoteMaster request.
   USE_SSL_CLIENT_AUTH	Whether to use SSL client authentication. Possible values are true, false. If true, you need to set the clientKey and clientCertificates fields in the demoteMaster request.

       mysql> call mysql.setupExternalSourceAutoPosition('1.1.1.1', 3306, \
       'USERNAME', 'PASSWORD', \
       /* master_auto_position= */ 1,false, false); \
   

   binlog

   Run the mysql.setupExternalSource(HOST, PORT, USER_NAME, USER_PASSWORD, SOURCE_LOG_NAME, SOURCE_LOG_POS, USE_SSL, USE_SSL_CLIENT_AUTH) stored procedure.

   Property	Description
   HOST	Source endpoint.
   PORT	Source port.
   USER_NAME	Source user.
   USER_PASSWORD	Source user password.
   SOURCE_LOG_NAME	The name of the binary log on the source database instance that contains the replication information.
   SOURCE_LOG_POS	The location in the mysql_binary_log_file_name binary log at which replication starts reading the replication information.
   USE_SSL	Whether to use SSL-based replication. Possible values are true, false. If true, you need to set the caCertificate field in the DemoteMaster request.
   USE_SSL_CLIENT_AUTH	Whether to use SSL client authentication. Possible values are true, false. If true, you need to set the clientKey and clientCertificates fields in the demoteMaster request.

       mysql> call mysql.setupExternalSource('1.1.1.1', 3306, \
       'user_name', 'password', 'mysql-bin-changelog.033877', 360, \
       false, false);
   

4. Use the mysql.startReplication() stored procedure to start replication from the external database.

      mysql> call mysql.startReplication();
   

5. Verify the replication status. Make sure that both the Slave_IO_Running and Slave_SQL_Running fields say YES.

      mysql> show slave status\G
   

   The output from this command looks similar to the following:

      *************************** 1. row ***************************
                      Slave_IO_State: Waiting for master to send event
                         Master_Host: 1.1.1.1
                         Master_User: user_name
                         Master_Port: 3306
                       Connect_Retry: 60
                     Master_Log_File: mysql-bin-changelog.000001
                 Read_Master_Log_Pos: 1
                      Relay_Log_File: relay-log.000002
                       Relay_Log_Pos: 1
               Relay_Master_Log_File: mysql-bin-changelog.000001
                    Slave_IO_Running: Yes
                   Slave_SQL_Running: Yes
                     Replicate_Do_DB:
                 Replicate_Ignore_DB:
                  Replicate_Do_Table:
              Replicate_Ignore_Table:
             Replicate_Wild_Do_Table:
         Replicate_Wild_Ignore_Table: mysql.%
                          Last_Errno: 0
                          Last_Error:
                        Skip_Counter: 0
                 Exec_Master_Log_Pos: 412
                     Relay_Log_Space: 752
                     Until_Condition: None
                      Until_Log_File:
                       Until_Log_Pos: 0
                  Master_SSL_Allowed: No
                  Master_SSL_CA_File:
                  Master_SSL_CA_Path:
                     Master_SSL_Cert:
                   Master_SSL_Cipher:
                      Master_SSL_Key:
               Seconds_Behind_Master: 0
       Master_SSL_Verify_Server_Cert: No
                       Last_IO_Errno: 0
                       Last_IO_Error:
                      Last_SQL_Errno: 0
                      Last_SQL_Error:
         Replicate_Ignore_Server_Ids:
                    Master_Server_Id: 1509941531
                         Master_UUID: 1cb2c80e-90f0-11eb-9ea3-02389b1c2e6f
                Master_Info_File: mysql.slave_master_info
                       SQL_Delay: 0
             SQL_Remaining_Delay: NULL
         Slave_SQL_Running_State: Slave has read all r
              Master_Retry_Count: 86400
                     Master_Bind:
         Last_IO_Error_Timestamp:
        Last_SQL_Error_Timestamp:
                  Master_SSL_Crl:
              Master_SSL_Crlpath:
              Retrieved_Gtid_Set:
               Executed_Gtid_Set: 478af53c-bd24-11eb-be72-42010a80002a:1-226
                   Auto_Position: 0
      1 row in set (0.00 sec)
   

Proceed with replication

Once you start replication from the external server, you need to monitor replication and then complete your migration. To learn more, see Monitoring replication.

Troubleshoot

Issue	Troubleshooting
Lost connection to MySQL server during query when dumping table.	The source may have become unavailable, or the dump contained packets too large. Make sure the external primary is available to connect. You can also modify the values of the net_read_timeout and net_write_timeout flags on the source instance to stop the error. For more information on the allowable values for these flags, see Configure database flags. To learn more about using mysqldump flags for managed import migration, see Allowed and default initial sync flags
The initial data migration was successful, but no data is being replicated.	One possible root cause could be your source database has defined replication flags which result in some or all database changes not being replicated over. Make sure the replication flags such as binlog-do-db, binlog-ignore-db, replicate-do-db or replicate-ignore-db are not set in a conflicting way. Run the command show master status on the primary instance to see the current settings.
The initial data migration was successful but data replication stops working after a while.	Things to try: Check the replication metrics for your replica instance in the Cloud Monitoring section of the Google Cloud console. The errors from the MySQL IO thread or SQL thread can be found in Cloud Logging in the mysql.err log files. The error can also be found when connecting to the replica instance. Run the command SHOW SLAVE STATUS, and check for the following fields in the output: Slave_IO_Running Slave_SQL_Running Last_IO_Error Last_SQL_Error
mysqld check failed: data disk is full.	The data disk of the replica instance is full. Increase the disk size of the replica instance. You can either manually increase the disk size or enable auto storage increase.

Review your replication logs

When you verify your replication settings, logs are produced.

You can view these logs by following these steps:

1. Go to the Logs Viewer in the Google Cloud console.

   Go to the Logs Viewer

2. Select the Cloud SQL replica from the Instance dropdown.
3. Select the replication-setup.log log file.

If the Cloud SQL replica is unable to connect to the external server, confirm the following:

• Any firewall on the external server is configured to allow connections from the Cloud SQL replica's outgoing IP address.
• Your SSL/TLS configuration is correct.
• Your replication user, host, and password are correct.

What's next

• Learn about updating an instance.
• Learn about managing replicas.
• Learn about monitoring instances.
• Learn about promoting your Cloud SQL replica to promote the replica to a standalone instance and stop replicating from the external server.