Docs Menu
Docs Home
/
MongoDB Atlas

atlas backups restores start

On this page

  • Syntax
  • Arguments
  • Options
  • Inherited Options
  • Output
  • Examples

Start a restore job for your project and cluster.

If you create an automated or pointInTime restore job, Atlas removes all existing data on the target cluster prior to the restore.

To use this command, you must authenticate with a user account or an API key with the Project Owner role. Atlas supports this command only for M10+ clusters.

Command Syntax
atlas backups restores start <automated|download|pointInTime> [options]
Name
Type
Required
Description
automated|download|pointInTime
string
true
Type of restore job to create. Valid values include: automated, download, pointInTime. To learn more about types of restore jobs, see https://www.mongodb.com/docs/atlas/backup-restore-cluster/.
Name
Type
Required
Description
--clusterName
string
true
-h, --help
false
help for start
--oplogInc
int
false
32-bit incrementing ordinal that represents operations within a given second. When paired with oplogTs, they represent the point in time to which your data will be restored.
--oplogTs
int
false
Oplog timestamp given as a timestamp in the number of seconds that have elapsed since the UNIX Epoch. When paired with oplogInc, they represent the point in time to which your data will be restored.
-o, --output
string
false
Output format. Valid values are json, json-path, go-template, or go-template-file. To see the full output, use the -o json option.
--pointInTimeUTCSeconds
int
false
Timestamp in the number of seconds that have elapsed since the UNIX epoch that represents the point in time to which your data will be restored. This timestamp must be within the last 24 hours of the current time.
--projectId
string
false
Hexadecimal string that identifies the project to use. This option overrides the settings in the configuration file or environment variable.
--snapshotId
string
false
Unique identifier of the snapshot to restore. You must specify a snapshotId for automated restores.
--targetClusterName
string
false
Name of the target cluster. For use only with automated restore jobs. You must specify a targetClusterName for automated restores.
--targetProjectId
string
false
Unique identifier of the project that contains the destination cluster for the restore job. You must specify a targetProjectId for automated restores.
Name
Type
Required
Description
-P, --profile
string
false
Name of the profile to use from your configuration file. To learn about profiles for the Atlas CLI, see https://dochub.mongodb.org/core/atlas-cli-save-connection-settings.

If the command succeeds, the CLI returns output similar to the following sample. Values in brackets represent your values.

Restore job '<Id>' successfully started
# Create an automated restore:
atlas backup restore start automated \
--clusterName myDemo \
--snapshotId 5e7e00128f8ce03996a47179 \
--targetClusterName myDemo2 \
--targetProjectId 1a2345b67c8e9a12f3456de7
# Create a point-in-time restore:
atlas backup restore start pointInTime \
--clusterName myDemo \
--pointInTimeUTCSeconds 1588523147 \
--targetClusterName myDemo2 \
--targetProjectId 1a2345b67c8e9a12f3456de7
# Create a download restore:
atlas backup restore start download \
--clusterName myDemo \
--snapshotId 5e7e00128f8ce03996a47179
What is  →