Snowflake Postgres point-in-time recovery¶

1. In the navigation menu, select Postgres.

2. Select the instance you want to fork.

3. Under Manage on the Postgres Instance page, select the Fork item and enter the configuration options.

   

4. Select Fork to create the fork.

To create a Postgres instance as a fork of an origin instance, execute the CREATE POSTGRES INSTANCE command and specify the FORK clause. The command creates the fork from the origin instance at the point in time specified by the AT or BEFORE clause. If you omit this clause, the fork is based on the origin instance at the current point in time.

CREATE POSTGRES INSTANCE <name>
  FORK <orig_name>
  [ { AT | BEFORE } ( { TIMESTAMP => <timestamp> | OFFSET => <time_difference> } ) ]
  [ COMPUTE_FAMILY = <compute_family> ]
  [ STORAGE_SIZE_GB = <storage_gb> ]
  [ HIGH_AVAILABILITY = { TRUE | FALSE } ]
  [ POSTGRES_SETTINGS = '<json_string>' ]
  [ COMMENT = '<string_literal>' ]
  [ [ WITH ] TAG ( <tag_name> = '<tag_value>' [ , ... ] ) ]

For the command parameters:

FORK orig_name

Specifies the origin of the fork.

{ AT | BEFORE } ( { TIMESTAMP => timestamp | OFFSET => time_difference } )

Specifies the point in time to fork from. The timestamp or offset must fall within the 10 day postgres data retention time.

Default: Uses current time.

The AT | BEFORE clause accepts one of the following parameters:

TIMESTAMP => timestamp

Specifies an exact date and time to use for Time Travel. The value must be explicitly cast to a TIMESTAMP, TIMESTAMP_LTZ, TIMESTAMP_NTZ, or TIMESTAMP_TZ data type.

If no explicit cast is specified, the timestamp in the AT clause is treated as a timestamp with the UTC time zone (equivalent to TIMESTAMP_NTZ). Using the TIMESTAMP data type for an explicit cast may also result in the value being treated as a TIMESTAMP_NTZ value. For details, see Date & time data types.

OFFSET => time_difference

Specifies the difference in seconds from the current time to use for Time Travel, in the form -N where N can be an integer or arithmetic expression (e.g. -120 is 120 seconds, -30*60 is 1800 seconds or 30 minutes).

Default: Copied from the origin.

COMPUTE_FAMILY = compute_family

Specifies the name of an instance size from the Snowflake Postgres Instance Sizes tables.

Default: Copied from the origin.

STORAGE_SIZE_GB = storage_gb

Specifies storage size in GB. Must be between 10 and 65,535.

Default: Copied from the origin.

HIGH_AVAILABILITY = { TRUE | FALSE }

Specifies the high availability setting to be used for the fork.

Default: Copied from the origin.

POSTGRES_SETTINGS = 'json_string'

Allows you to optionally set Postgres configuration parameters on your instance in JSON format. See Snowflake Postgres Server Settings for a list of available Postgres parameters.

'{"component:name" = "value", ...}'

Copy

Default: Copied from the origin.

COMMENT = 'string_literal'

Specifies a comment for the user.

Default: NULL

TAG ( tag_name = 'tag_value' [ , tag_name = 'tag_value' , ... ] )

Specifies the tag name and the tag string value.

The tag value is always a string, and the maximum number of characters for the tag value is 256.

For information about specifying tags in a statement, see Tag quotas.

One row with the following columns will be returned:

• status

• host

CREATE FORK SQL Examples

Create a fork my_fork from the origin instance my_origin_instance at the timestamp 2025-01-01 12:00:00.

CREATE POSTGRES INSTANCE my_fork
  FORK my_origin_instance
  AT (TIMESTAMP => '2025-01-01 12:00:00');

Copy

Create a fork my_fork from the origin instance my_origin_instance as it was 120 seconds ago.

CREATE POSTGRES INSTANCE my_fork
  FORK my_origin_instance
  AT (OFFSET => -120);

Copy

Create a fork my_fork from the origin instance my_origin_instance as of the current time, using the STANDARD_M instance size and no high availablility.

CREATE POSTGRES INSTANCE my_fork
  FORK my_origin_instance
  COMPUTE_FAMILY = STANDARD_M
  HIGH_AVAILABILITY = FALSE;

Copy