SingleStore DB



This command is in preview mode. At a future date, SingleStore will announce when this command will be available for production use.

Prior to using this command, the enable_bottomless and allow_point_in_time_attach engine variables must be set to ON. The default values for enable_bottomless and allow_point_in_time_attach, respectively, are ON and OFF.

This command is not available in the SingleStore DB Free Edition.

Restores an unlimited storage database at a restore point and then brings the database online.

Check here for the permissions required to use this command.

ATTACH DATABASE remote_database_name ON <object_store_settings> [AT MILESTONE milestone_name | AT TIME attach_time]


    S3 { '<bucket-name>' | '<bucket-name/object-name>' | '<bucket-name/prefix/object-name>' }
      [CONFIG '<configuration_json>']
      CREDENTIALS '<credentials_json>'

A DATETIME or DATETIME(6) literal.


A quoted string, which is the name of a milestone to restore to.

  • If remote_database_name is currently attached, it must be detached (using DETACH DATABASE) before it can be reattached.

  • AT TIME attaches the database, with its contents restored to the specified point in time. The specified time may be a DATETIME or a DATETIME(6) literal. An error will be returned if you attempt to attach a database at time literal that is later than the time of the latest detach.

  • If AT MILESTONE and AT TIME are not specified, the database is attached at the latest state where all information is available for committed transactions.

  • If AT MILESTONE or AT TIME are specified, the restore point must be within the unlimited storage database retention period.

  • You cannot specify both AT MILESTONE and AT TIME.

  • remote_database_name of the attached database can be different than the remote_database_name used when the database was detached.

  • Attaching an unlimited storage database can be faster than restoring an equivalent local storage database. This is because an attach of an unlimited storage database does not copy all data to the cluster, as is the case with the restore of a local storage database. Note that after an unlimited storage database is attached, queries may be slower for some time until remote data is cached locally in the cluster.