Use a managed import to set up replication from external databases

MySQL   |  PostgreSQL   |  SQL Server

This page describes how to set up and use a managed import for data when replicating from an external server to Cloud SQL.

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

Before you begin

Before you begin, complete these steps:

1. Configure the external server.

2. Create the source representation instance.

3. 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 (%). Update this user account so that it can be used only with the Cloud SQL replica.

If the replication is based on GTID, you need the REPLICATION SLAVE, EXECUTE, SELECT, SHOW VIEW, REPLICATION CLIENT, and RELOAD permissions.

If the replication is based on binlog, you need the REPLICATION SLAVE, EXECUTE, SELECT, SHOW VIEW, REPLICATION CLIENT, and RELOAD permissions.

Open a terminal on the external server and enter the following commands.

mysql Client

For GTID:

    UPDATE mysql.user
      SET Host='NEW_HOST'
      WHERE Host='OLD_HOST'
      AND User='USERNAME';
    GRANT REPLICATION SLAVE, EXECUTE, SELECT, SHOW VIEW, REPLICATION_CLIENT,
    RELOAD ON . TO
    'USERNAME'@'HOST';
    FLUSH PRIVILEGES;

For binlog:

    UPDATE mysql.user
    SET Host='NEW_HOST'
    WHERE Host='OLD_HOST'
    AND User='USERNAME';
    GRANT REPLICATION SLAVE, EXECUTE, SELECT, SHOW VIEW, REPLICATION CLIENT,
    RELOAD 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, SELECT, SHOW VIEW, REPLICATION CLIENT,
RELOAD ON *.* TO 'username'@'host.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.

Verify your replication settings

After your setup is complete, ensure that the Cloud SQL replica can replicate from the external server.

The following external sync settings must be correct.

• Connectivity between the Cloud SQL replica and external server
• Replication user privileges
• Version compatibility
• The Cloud SQL replica is not already replicating
• Binlogs are enabled on the external server
• GTID is enabled if you are trying to do an external sync from an RDS external server and are using a Google Cloud bucket

To verify these settings, open a Cloud Shell terminal and enter the following commands:

curl

gcloud auth login
ACCESS_TOKEN="$(gcloud auth print-access-token)"
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
     --header 'Content-Type: application/json' \
     --data '{
         "syncMode": "SYNC_MODE",
         "mysqlSyncConfig": {
           "initialSyncFlags": "SYNC_FLAGS"
         }
       }' \
     -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/PROJECT_ID/instances/REPLICA_INSTANCE/verifyExternalSyncSettings

example

gcloud auth login
ACCESS_TOKEN="$(gcloud auth print-access-token)"
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
     --header 'Content-Type: application/json' \
     --data '{
         "syncMode": "online"
       }' \
     -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/myproject/instances/myreplica/verifyExternalSyncSettings

example w/ sync flags

gcloud auth login
ACCESS_TOKEN="$(gcloud auth print-access-token)"
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
     --header 'Content-Type: application/json' \
     --data '{
         "syncMode": "online",
         "mysqlSyncConfig": {
             "initialSyncFlags": [{"name": "max-allowed-packet", "value": "1073741824"}, {"name": "hex-blob"}, {"name": "compress"}]
             }
       }' \
     -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/MyProject/instances/replica-instance/verifyExternalSyncSettings

These calls return a list of type sql#externalSyncSettingErrorList.

If the list is empty, there are no errors. A response without errors appears like this: { "kind": "sql#externalSyncSettingErrorList" }

Property	Description
SYNC_MODE	Ensures that you can keep the Cloud SQL replica and the external server in sync after replication is set up. Sync modes include EXTERNAL_SYNC_MODE_UNSPECIFIED, ONLINE, and OFFLINE.
SYNC_FLAGS	A list of initial sync flags to verify. Only recommended if you plan on using custom sync flags when replicating from the source. For a list of allowed flags see here.
PROJECT_ID	The ID of your project in Google Cloud.
REPLICA_INSTANCE	The ID of your Cloud SQL replica.

Global read lock permission

If you don't have permission to access the global read lock on the external server, as might be the case with Amazon RDS and Amazon Aurora, pause writes to your server as described in the following steps:

1. Go to the Logs Explorer, and select your Cloud SQL replica from the resource list. You should see a list of the most recent logs for your Cloud SQL replica. Ignore them for now.
2. Open a terminal and enter the commands in Start replication on the external server to replicate from the external server.

3. Return to the Logs Explorer. When you see the log as follows, stop writing to the database on your external server. In most cases, this is required only for a few seconds.

      DUMP_IMPORT(START): Start importing data, please pause any write to the
      external primary database.
   

4. When you see the following log entry in Logs Explorer, re-enable writing to the database on your external server.

      DUMP_IMPORT(SYNC): Consistent state on primary and replica. Writes to the
      external primary may resume.
   
   

Start replication on the external server

After you have verified that you can replicate from the external server, you are ready to perform the replication. Expect the replica to import about 25-50 GB per hour.

Caution: When you initiate an import from a database on a managed service, such as Amazon RDS or Amazon Aurora, Cloud SQL instructs you to stop writes for a few seconds so that the data dump is consistent. To avoid downtime for writes to your managed service primary instance, you can select a read replica as your replication source instead of the primary instance. During the phase when stopping writes is required, you can temporarily stop replication to the source read replica and then resume replication once the stop writes phase is over.

During the initial import process, do not perform any DDL operations on the external server. Doing so could cause inconsistencies during the import. After the import process completes, the replica uses the binary logs on the external server to catch up to the current state of the external server.

curl

gcloud auth login
ACCESS_TOKEN="$(gcloud auth print-access-token)"
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
     --header 'Content-Type: application/json' \
     --data '{
         "syncMode": "SYNC_MODE",
         "skipVerification": "SKIP_VERIFICATION",
         "mysqlSyncConfig": {
           "initialSyncFlags": "SYNC_FLAGS"
         }
       }' \
     -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/PROJECT_ID/instances/REPLICA_INSTANCE/startExternalSync

example

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

example w/ sync flags

gcloud auth login
ACCESS_TOKEN="$(gcloud auth print-access-token)"
curl --header "Authorization: Bearer ${ACCESS_TOKEN}" \
     --header 'Content-Type: application/json' \
     --data '{
         "syncMode": "online",
         "skipVerification": false,
         "mysqlSyncConfig": {
             "initialSyncFlags": [{"name": "max-allowed-packet", "value": "1073741824"}, {"name": "hex-blob"}, {"name": "compress"}]
             }
       }' \
     -X POST \
     https://sqladmin.googleapis.com/sql/v1beta4/projects/MyProject/instances/replica-instance/startExternalSync

Property	Description
SYNC_MODE	Verifies that you can keep the Cloud SQL replica and external server in sync after replication is set up.
SKIP_VERIFICATION	Whether or not to skip the built-in verification step before syncing your data. Only recommended if you have already verified your replication settings.
SYNC_FLAGS	A list of initial sync flags to verify. Only recommended if you plan on using custom sync flags when replicating from the source. For a list of allowed flags see here.
PROJECT_ID	The ID of your project in Google Cloud.
REPLICA_INSTANCE	The ID of your Cloud SQL replica.

Allowed and default initial sync flags

When setting up replication when using managed import, mysqldump runs with the following flags by default:

• --host=[HOST_SPECIFIED_IN_SOURCE_CONFIG]
• --port=[PORT_SPECIFIED_IN_SOURCE_CONFIG]
• --ssl-ca=[SSL_CA_SPECIFIED_IN_SOURCE_CONFIG]
• --ssl-cert=[SSL_CERT_SPECIFIED_IN_SOURCE_CONFIG]
• --ssl-key=[SSL_KEY_SPECIFIED_IN_SOURCE_CONFIG]
• --master-data=[SPECIFIED_BY_SYNC_MODE_AND_SOURCE_GTID_MODE]
• --set-gtid-purged=AUTO
• --single-transaction
• --hex-blob
• --compress
• --no-autocommit
• --default-character-set=utf8mb4
• --databases [ALL_NON_SYSTEM_DBS]
• --add-drop-table
• --routines

We recommend this configuration for most customers.

However, for particular cases, you might need to specify custom sync flags. The following flags are allowed:

• --add-drop-database
• --add-drop-table
• --add-drop-trigger
• --add-locks
• --allow-keywords
• --all-tablespaces
• --apply-slave-statements
• --column-statistics
• --comments
• --compact
• --compatible
• --complete-insert
• --compress
• --compression-algorithms
• --create-options
• --default-character-set
• --delayed-insert
• --disable-keys
• --dump-date
• --events
• --extended-insert
• --fields-enclosed-by
• --fields-escaped-by
• --fields-optionally-enclosed-by
• --fields-terminated-by
• --flush-logs
• --flush-privileges
• --force
• --get-server-public-key
• --hex-blob
• --ignore-error
• --ignore-read-lock-error
• --ignore-table
• --insert-ignore
• --lines-terminated-by
• --lock-all-tables
• --lock-tables
• --max-allowed-packet
• --net-buffer-length
• --network-timeout
• --no-autocommit
• --no-create-db
• --no-create-info
• --no-data
• --no-defaults
• --no-set-names
• --no-tablespaces
• --opt
• --order-by-primary
• --pipe
• --quote-names
• --quick
• --replace
• --routines
• --secure-auth
• --set-charset
• --shared-memory-base-name
• --show-create-skip-secondary-engine
• --skip-opt
• --ssl-cipher
• --ssl-fips-mode
• --ssl-verify-server-cert
• --tls-ciphersuites
• --tls-version
• --triggers
• --tz-utc
• --verbose
• --xml
• --zstd-compression-level

For allowed values, see the MySQL public docs.

Monitor the migration

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

Troubleshoot

Consider the following troubleshooting options:

Issue	Troubleshooting
Read replica didn't start replicating on creation.	There's probably a more specific error in the log files. Inspect the logs in Cloud Logging to find the actual error.
Unable to create read replica - invalidFlagValue error.	One of the flags in the request is invalid. It could be a flag you provided explicitly or one that was set to a default value. First, check that the value of the max_connections flag is greater than or equal to the value on the primary. If the max_connections flag is set appropriately, inspect the logs in Cloud Logging to find the actual error.
Unable to create read replica - unknown error.	There's probably a more specific error in the log files. Inspect the logs in Cloud Logging to find the actual error. If the error is: set Service Networking service account as servicenetworking.serviceAgent role on consumer project, then disable and re-enable the Service Networking API. This action creates the service account necessary to continue with the process.
Disk is full.	The primary instance disk size can become full during replica creation. Edit the primary instance to upgrade it to a larger disk size.
The replica instance is using too much memory.	The replica uses temporary memory to cache often-requested read operations, which can lead it to use more memory than the primary instance. Restart the replica instance to reclaim the temporary memory space.
Replication stopped.	The maximum storage limit was reached and automatic storage increase isn't enabled. Edit the instance to enable automatic storage increase.
Replication lag is consistently high.	The write load is too high for the replica to handle. Replication lag takes place when the SQL thread on a replica is unable to keep up with the IO thread. Some kinds of queries or workloads can cause temporary or permanent high replication lag for a given schema. Some of the typical causes of replication lag are: Slow queries on the replica. Find and fix them. All tables must have a unique/primary key. Every update on such a table without a unique/primary key causes full table scans on th replica. Queries like DELETE ... WHERE field < 50000000 cause replication lag with row-based replication since a huge number of updates are piled up on the replica. Some possible solutions include: Configure parallel replication. Set the innodb_flush_log_at_trx_commit flag on the read replica to 2. See Tips for working with flags for more information about this flag. Edit the instance to increase the size of the replica. Reduce the load on the database. Send read traffic to the read replica. Index the tables. Identify and fix slow write queries. Recreate the replica.
Replication lag suddenly spikes.	This is caused by long-running transaction(s). When a transaction (single statement or multi-statements) commits on the source instance, the start time of the transaction is recorded in the binary log. When the replica receives this binlog event, it compares that timestamp with the current timestamp to calculate replication lag. Hence, a long-running transaction on the source would result in an immediate large replication lag on the replica. If the amount of row changes in the transaction is large, the replica would also spend a long time to execute it. During the time, replication lag is increasing. Once the replica finishes this transaction, the catch up period would depend on the write workload on the source and the replica's processing speed. To avoid a long transaction, some possible solutions include: Break the transaction into multiple small transactions Chunk a single large write query into smaller batches Try to separate long SELECT queries from a transaction mixed with DMLs
Changing parallel replication flags results in an error.	An incorrect value is set for one of or more of these flags. On the primary instance that's displaying the error message, set the parallel replication flags: Modify the binlog_transaction_dependency_tracking and transaction_write_set_extractionflags: binlog_transaction_dependency_tracking=COMMIT_ORDER transaction_write_set_extraction=OFF Add the slave_pending_jobs_size_max flag: slave_pending_jobs_size_max=33554432 Modify the transaction_write_set_extraction flag: transaction_write_set_extraction=XXHASH64 Modify the binlog_transaction_dependency_tracking flag: binlog_transaction_dependency_tracking=WRITESET
Replica creation fails with timeout.	Long-running uncommitted transactions on the primary instance can cause read replica creation to fail. Recreate the replica after stopping all running queries.

Additionally, for MySQL, also consider the following options:

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, or use mysqldump with the max_allowed_packet option. 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.