autorestic

High backup level CLI utility for restic.

Autorestic is a wrapper around the amazing restic. While being amazing the restic cli can be a bit overwhelming and difficoult to manage if you have many different location that you want to backup to multiple locations. This utility is aimed at making this easier 🙂

🌈 Features

• Config files, no CLI
• Predictable
• Backup locations to multiple backends
• Snapshot policies and pruning
• Simple interface
• Fully encrypted

Installation

curl -s https://raw.githubusercontent.com/CupCakeArmy/autorestic/master/install.sh | bash

🚀 Quickstart

Setup

First we need to configure our locations and backends. Simply create a .autorestic.yml either in your home directory of in the folder from which you will execute autorestic.

Optionally you can specify the location of your config file by passing it as argument: autorestic -c ../path/config.yml

locations:
  home:
    from: /home/me
    to: remote

  important:
    from: /path/to/important/stuff
    to:
      - remote
      - hdd

backends:
  remote:
    type: b2
    path: 'myBucket:backup/home'
    B2_ACCOUNT_ID: account_id
    B2_ACCOUNT_KEY: account_key

  hdd:
    type: local
    path: /mnt/my_external_storage

Then we check if everything is correct by running the check command. We will pass the -a (or --all) to tell autorestic to check all the locations.

Lets see a more realistic example (from the config above)

autorestic check -l important

If we would check only one location we could run the following: autorestic -l home check.

Backup

autorestic backup -a

Restore

autorestic restore -a --to /path/where/to/restore

This will restore all the locations to the selected target. If for one location there are more than one backends specified autorestic will take the first one.

autorestic restore -l home --from hdd --to /path/where/to/restore

This will restore the location home to the /path/where/to/restore folder and taking the data from the backend hdd

🗂 Locations

A location simply a folder on your machine that restic will backup. The paths can be relative from the config file. A location can have multiple backends, so that the data is secured across multiple servers.

locations:
  my-location-name:
    from: path/to/backup
    to:
      - name-of-backend
      - also-backup-to-this-backend

💽 Backends

Backends are the place where you data will be saved. Backups are incremental and encrypted.

Fields

type

Type of the backend see a list here

Supported are:

• Local
• Backblaze B2
• Amazon S3
• Minio
• Google Cloud Storage
• Microsoft Azure Storage

For each backend you need to specify the right variables as shown in the example below.

path

The path on the remote server. For object storages as

Example

backends:
  name-of-backend:
    type: b2
    path: 'myAccount:myBucket/my/path'
    B2_ACCOUNT_ID: backblaze_account_id
    B2_ACCOUNT_KEY: backblaze_account_key

Pruning and snapshot policies

Autorestic supports declaring snapshot policies for location to avoid keeping old snapshot around if you don't need them.

This is based on Restic's snapshots policies, and can be enabled for each location as shown below:

locations:
  etc:
    from: /etc
    to: local
    keep:
      # options matches the --keep-* options used in the restic forget CLI
      # cf https://restic.readthedocs.io/en/latest/060_forget.html#removing-snapshots-according-to-a-policy
      last: 5             # always keep at least 5 snapshots
      hourly: 3           # keep 3 last hourly shapshots
      daily: 4            # keep 4 last daily shapshots
      weekly: 1           # keep 1 last weekly shapshots
      monthly: 12         # keep 12 last monthly shapshots
      yearly: 7           # keep 7 last yearly shapshots
      within: "2w"        # keep snapshots from the last 2 weeks

Pruning can be triggered using autorestic forget -a, for all locations, or selectively with autorestic forget -l <location>. please note that contrary to the restic CLI, restic forget will call restic prune internally.

Run with the --dry-run flag to only print information about the process without actually pruning the snapshots. This is especially useful for debugging or testing policies:

$ autorestic forget -a --dry-run --verbose

Configuring Backends
local : Done ✓

Removing old shapshots according to policy
etc ▶ local : Removing old spnapshots… ⏳
etc ▶ local : Running in dry-run mode, not touching data
etc ▶ local : Forgeting old snapshots… ⏳Applying Policy: all snapshots within 2d of the newest
keep 3 snapshots:
ID        Time                 Host        Tags        Reasons    Paths
-----------------------------------------------------------------------------
531b692a  2019-12-02 12:07:28  computer                within 2w  /etc
51659674  2019-12-02 12:08:46  computer                within 2w  /etc
f8f8f976  2019-12-02 12:11:08  computer                within 2w  /etc
-----------------------------------------------------------------------------
3 snapshots

Note

Note that the data is automatically encrypted on the server. The key will be generated and added to your config file. Every backend will have a separate key. You should keep a copy of the keys somewhere in case your server dies. Otherwise DATA IS LOST!

Contributors

This amazing people helped the project!

• @ChanceM [Docs]
• @EliotBerriot [Docs, Pruning, S3]