On this page Carat arrow pointing down

This feature is in preview. This feature is subject to change. To share feedback and/or issues, contact Support.

New in v22.2: You can use external connections to specify and interact with resources that are external from CockroachDB. With CREATE EXTERNAL CONNECTION, you define a name for an external connection while passing the provider URI and query parameters. BACKUP, RESTORE, IMPORT, EXPORT, and CREATE CHANGEFEED queries can interact with the defined external connection instead of a required, provider-specific URI. As a result, you can decouple the management of the external resource from the operation in which you're using them.

CREATE EXTERNAL CONNECTION will validate the URI by writing, reading, and listing a test file to the external storage URI. If you're using a KMS URI, CREATE EXTERNAL CONNECTION will encrypt and decrypt a file. You'll find a crdb_external_storage_location file in your external storage as a result of this test. Each of the operations that access the external connection is aware of the raw URI that is parsed to configure, authenticate, and interact with the connection.

The privilege model for external connections means that you can delegate the creation and usage of external connections to the necessary users or roles.

You can also use the following SQL statements to work with external connections:

Required privileges

To create an external connection, a user must have the EXTERNALCONNECTION system-level privilege. root and admin users have this system-level privilege by default and are capable of granting the EXTERNALCONNECTION system-level privilege to other users and roles with or without the GRANT OPTION.

For example:


To use a specific external connection during an operation, the user must also have the USAGE privilege on that connection:

For example:



CREATE EXTERNAL CONNECTION connection_name AS connection_URI


Parameter Description
connection_name The name that represents the external connection.
connection_URI The storage/sink URI that the external connection will reference.

Supported external storage and sinks

Storage or sink Operation support
Amazon S3 Backups, restores, imports, exports
Amazon S3 KMS Encrypted backups
Azure Storage Backups, restores, imports, exports
Google Cloud Storage Backups, restores, imports, exports
Google Cloud Storage KMS Encrypted backups
Kafka Changefeeds
Nodelocal Backups, restores, imports, exports
Userfile Backups, restores, imports, exports

For more information on authentication and forming the URI that an external connection will represent, see each of the links to the storage or sink pages in the table.

Changefeed sinks as external connections

Consider the following when you create an external connection for:

  • Kafka sinks: You can only include the query parameters and options that Kafka sinks support. See the Options table and the Kafka query parameters for more detail.
  • Cloud storage sinks: You must include the key_in_value option in the changefeed creation statement if you are emitting messages to a cloud storage sink external connection.

External connection URI format

To form the external connection URI in operation statements, use the external:// scheme followed by the name of the external connection.

For an external connection named backup_storage:


See the examples in the next section for details on external connection creation.


The examples in this section demonstrate some of the storage and operation options that external connections support.

Create an external connection for cloud storage

In this example, you create an external connection for an Amazon S3 bucket that will store your backups. Then, you use the external connection to restore the backup to your cluster.

  1. Define your external connection that references the S3 bucket's URI:

    CREATE EXTERNAL CONNECTION backup_bucket AS 's3://bucket name?AWS_ACCESS_KEY_ID={access key}&AWS_SECRET_ACCESS_KEY={secret access key}';
  2. Verify that the new external connection was created successfully with SHOW CREATE EXTERNAL CONNECTION:

    connection_name |                                                                              create_statement
    backup_bucket   | CREATE EXTERNAL CONNECTION 'backup_bucket' AS 's3://bucket name?AWS_ACCESS_KEY_ID={access key}&AWS_SECRET_ACCESS_KEY={secret access key}'
  3. Run the backup to your S3 bucket using the external connection's name:

    BACKUP DATABASE movr INTO 'external://backup_bucket' AS OF SYSTEM TIME '-10s' WITH revision_history;
  4. Use SHOW BACKUP to view your backups in the storage defined by the external connection:

    SHOW BACKUPS IN 'external://backup_bucket';
    (2 rows)
  5. In the event that a restore is necessary, use RESTORE with the external connection:

    RESTORE DATABASE movr FROM LATEST IN 'external://backup_bucket';
  6. When you no longer need the external connection, you can delete it with DROP EXTERNAL CONNECTION:


Create an external connection for a changefeed sink

In this example, you create an external connection to a Kafka sink to which a changefeed will emit messages. When you create the external connection, you will include the necessary query parameters for your changefeed. As a result, you will only need to specify the external connection's name when creating a changefeed rather than the Kafka URI and parameters.

  1. Define your external connection that references the Kafka sink URI and any query parameters:

    CREATE EXTERNAL CONNECTION kafka_sink AS 'kafka://{certificate}&sasl_enabled=true&sasl_user={sasl user}&sasl_password={url-encoded password}&sasl_mechanism=SASL-SCRAM-SHA-256';
  2. Create your changefeed using the external connection's name:

    CREATE CHANGEFEED FOR TABLE movr.users INTO 'external://kafka_sink' WITH resolved;

See also

Yes No
On this page

Yes No