Skip to main content
Rclone
Last update:

Rclone

Rclone — is a command line interface for working with objects in object storage.

Customize Rclone⁠

  1. Configure access.
  2. Install the client.
  3. Create an Rclone configuration.

Set up access⁠

Access can be configured by a user with role Account Owner or User Administrator.

  1. Create a service user с a role with access to object storage. If you are using a user with the Object Storage User role, the container must be configured with the access policy.
  2. Issue an S3 key to the user.

Install the client⁠

  1. On Rclone website download the distribution for Windows.

  2. Unpack the archive with the distribution.

  3. Open a command prompt.

  4. Navigate to the directory with the distribution:

    cd <path>

    Specify <path> — path to the directory with the distribution.

Create an Rclone configuration⁠

  1. Open the configuration mode:

    rclone config

    A dialog box will appear in the console:

    No remotes found, make a new one?
    n) New remote
    s) Set configuration password
    q) Quit config

  2. Create a new configuration:

    n/s/q> n

  3. Click Enter.

  4. Enter the name of the remote storage connection (for example, selectel_s3):

    name> selectel_s3

  5. Click Enter.

  6. Enter the storage type that is specified after the string Amazon S3 Compliant Storage Provider (AWS, Alibaba, Ceph, Digital Ocean, Dreamhost, IBM COS, Minio, etc):

    Storage> s3

  7. Click Enter.

  8. Enter the type of provider that is specified after the line Any other S3 compatible provider:

    provider> Other

  9. Click Enter.

  10. Select the method of obtaining authorization data that is specified after the line Enter AWS credentials in the next step:

    env_auth> false

  11. Enter AWS Access Key ID:

    access_key_id> <access_key>

    Specify <access_key> — field value Access key from S3 key.

  12. Enter AWS Secret Access Key (password):

    secret_access_key> <secret_key>

    Specify <secret_key> — field value Secret key from S3 key.

  13. Enter pool, in which the object store is located (e.g., ru-1):

    region> <pool>

  14. Enter the URL for S3 API access:

    endpoint> <s3_domain>

    Specify <s3_domain> — S3 API domain depends on pool where the object storage is located.

  15. Parameter location_constraint leave it blank.

  16. Click Enter.

  17. Parameter acl leave it blank.

  18. Click Enter.

  19. Parameter Edit advanced config leave blank or enter n.

  20. Click Enter.

  21. Check the configuration:

    Remote config
    --------------------
    [selectel_s3]
    type = s3
    provider = Other
    env_auth = false
    access_key_id = <access_key>
    secret_access_key = <secret_key>
    region = <pool>
    endpoint = <s3_domain>
    --------------------
    y) Yes this is OK
    e) Edit this remote
    d) Delete this remote

  22. Confirm the creation of the configuration:

    y/e/d> y

  23. Click Enter.

  24. Exit the settings dialog box:

    e/n/d/r/c/s/q> q

  25. Click Enter.

Working with Rclone⁠

Command format⁠

General view of the teams:

<command> <remote_name>:<container_name>/<object_name>

Specify:

  • <command> — an Rclone command of the form rclone <subcommand>. You can view all the teams at Rclone website or in the output of the command rclone --help
  • <remote_name> — the connection name specified when configuring Rclone;
  • <container_name> — container name;
  • <object_name> — optional: object name.

Create a container⁠

Only a private container can be created via the S3 API.

Create a container:

rclone mkdir <remote_name>:<container_name>

Specify:

  • <remote_name> — the connection name specified when configuring Rclone;
  • <container_name> — the name of the new container. Observe the rules for naming containers, see the instructions for more details Bucket naming rules AWS documentation.

View list of objects⁠

Check out the list of facilities:

rclone ls <remote_name>:<container_name>/<prefix>

Specify:

  • <remote_name> — the connection name specified when configuring Rclone;
  • <container_name> — container name;
  • <prefix> — optional: prefix the objects you want to output.

Load object⁠

When performing copy and synchronization operations, Rclone checks all objects by date and time of change or md5 sum. Objects that have been modified are transferred from the source directory to the destination directory.

Load the object from the local directory:

rclone copy <path> <remote_name>:<container_name>/<object_name>

Specify:

  • <path> — path to the object in the local directory;
  • <remote_name> — the connection name specified when configuring Rclone;
  • <container_name> — container name;
  • <object_name> — optional: name with which the object will be loaded. If it is not specified, the file name will be used.

Delete object⁠

Delete the object:

rclone deletefile <remote_name>:<container_name>/<object_name>

Specify:

  • <remote_name> — the connection name specified when configuring Rclone;
  • <container_name> — container name;
  • <object_name> — the name of the object to be deleted.