This call requires the path to a backup, and then restores the Tamr Core instance to that backup. Acceptable backup locations are S3, HDFS, GCS, or a local folder. See Get all backups or Get a backup by ID.
For instructions on how to restore single-node Tamr Core deployments, see Restore.
Note: If the backup is in S3, before you start the restore operation, specify your access credentials with the following configuration parameters:
TAMR_UNIFY_BACKUP_AWS_SECRET_ACCESS_KEY. See AWS Backup and Restore.
After Tamr Core starts the restore process, the service returns a JSON object with the same content as a backup. See Get all backups. You can use the relativeID, such as
restore-2022-01-14_18-44-54-084, to look up the status of the restore operation. See Get the description of the most recent restore operation, if any.
The restore process puts Tamr Core in a quiet state where only the restore process runs, most service endpoints are unavailable, and you cannot use the user interface.
The restore process reads backup data from the location you provide to the start restore service method. You can specify physical disk backup locations in AWS S3, GCS, and HDFS.
The restore process updates Tamr Core with a subset of the service-related configuration stored through the Tamr Core administration utility. See Additional Configuration Variables for Backup.
Any configuration settings that you have not stored with the backup or included in the list of automatically-backed up variables are overwritten during the restore operation. You can add all additional configuration properties to the
TAMR_UNIFY_BACKUP_EXTRA_CONFIG_PROPS list before you start the backup.
Tamr Core becomes available after the restore process completes. During the restore process, the error message, "The system is temporarily in read-only mode" appears.
When the restore process completes, you must restart Tamr Core and its dependencies to apply the new configuration. See Restarting.