Skip to main content

SAVE

Syntax​

SAVE [RDB|DF] [cloud_uri] [filename]

Time complexity: O(N) where N is the total number of keys in all databases

ACL categories: @admin, @slow, @dangerous

The SAVE commands performs a save of the dataset producing a point in time snapshot of all the data inside the Dragonfly instance, in the form of a set of DFS files.

Use SAVE RDB to save the snapshot in form of an RDB file instead. Please refer to the persistence documentation for detailed information about RDB files.

If cloud_uri argument is set, save will be done on remote cloud storage. Currently, SAVE command supports writing to S3 (argument starts with s3://) and GCP (argument starts with gs://).

Optional argument filename is used to override current dbfilename for saving file(s).

Flags​

  • df_snapshot_format - Set true to save dump in Dragonfly file format (true by default).
  • dbfilename - The file name to save and load the database. To generate a file with a timestamp, set the macro {timestamp} in the filename, e.g. dump-{timestamp}. The macro will be replaced with a timestamp of the local time in a lexicographically sorted format. The default filename is dump-{timestamp}.
  • snapshot_cron - Generate snapshots periodically. The argument is a cron format (e.g. 0 0 * * *, */5 * * * *)

Return​

Simple string reply: The commands returns OK on success.

Examples​

Save into current working directory with default file name.

dragonfly> SAVE DF
OK

Save into current working directory with my_dump as file name for saved file(s).

dragonfly> SAVE DF my_dump
OK

Save into S3 bucket and use my_dump as file name for file(s).

dragonfly> SAVE DF s3://bucket/folder my_dump
OK