duplyvolume

A simple solution to back up docker volumes.

Features:

• Supports encryption, incremental backups, and S3 storage based on duplicity
• Auto-discovery of docker volumes
• Automatically stop/start only the containers that use a volume
• Schedule backups using cron expressions
• Overwrite global retention period using volume labels

Not implemented:

• Backup/Restore of a single volume. You can use the duplicity command line for that.
• Better stop/start strategies. With the current strategy, a container can be stopped multiple times if it has multiple attached volumes or shares a volume with another container. This can lead to inconsistencies in the backup.

Example

Example docker-compose.yaml configuration:

services:
  duplyvolume:
    image: "ghcr.io/swnf/duplyvolume:1-latest"
    restart: unless-stopped
    environment:
      BACKUP_CRON: "0 3 * * 0"
      TZ: "Europe/Berlin"
    volumes:
      - "/var/run/docker.sock:/var/run/docker.sock"
      - "/path/to/backup:/target"

For S3 support specify S3_BUCKET_NAME/S3_REGION_CODE/AWS_ACCESS_KEY_ID/AWS_SECRET_ACCESS_KEY and remove the /target volume.

Commands

Command	Description
docker-compose exec duplyvolume backup	Perform a backup of all volumes (Press Ctrl-C to cancel)
docker-compose exec duplyvolume restore	Restore all volumes. This will overwrite all contents of your volumes. If a volume has no recent backups, duplyvolume will assume that it was deleted and will not restore it. (Press Ctrl-C to cancel)
docker-compose exec duplyvolume cancel	Cancel a running backup running somewhere else
docker-compose exec duplyvolume healthcheck	Perform a healthcheck
docker-compose stop duplyvolume	Cancel all running backups and shut down

Environment variables

All environment variable values can be substituted with files. Just point for example PASSPHRASE_FILE to the path of a file

Variable	Description
IGNORE_REGEX	Ignore volumes with names matching this regex. By default, volume names containing "tmp", "cache" and anonymous volumes are ignored.
BACKUP_CRON	A cron expression in the format year - month - day - week - day of week - hour - minute - second. "*" is the wildcard character. For more information, see here.
FULL_IF_OLDER_THAN	If the last backup is older than this timespan, perform a full instead of an incremental backup. Defaults to one month ("1M", see Time Formats).
REMOVE_OLDER_THAN	Delete all backups older than this timespan. Dependencies of newer backups will not be deleted.
REMOVE_ALL_BUT_N_FULL	Delete all backups older than the last n full backups.
REMOVE_ALL_INC_OF_BUT_N_FULL	Delete incremental backups older than the last n full backups.
TZ	The timezone used for the backup scheduler.
PASSPHRASE	The passphrase used to encrypt the backup. It will only encrypt volume contents, not volume metadata. If this is not set, the backup will be unencrypted.
S3_BUCKET_NAME	The name of a S3 bucket. If this is set, the bucket will be used instead of /target. This setting also requires S3_REGION_CODE or S3_ENDPOINT_URL.
S3_REGION_CODE	The region code of a S3 bucket. This setting also requires S3_BUCKET_NAME. It is mutually exclusive with S3_ENDPOINT_URL.
S3_ENDPOINT_URL	The endpoint url of a S3 bucket. This setting also requires S3_BUCKET_NAME. It is mutually exclusive with S3_REGION_CODE. Use this setting if you want to use a custom S3 compatible storage server.
S3_STORAGE_CLASS	The S3 storage class to use. Can only be STANDARD or STANDARD_IA. Defaults to STANDARD
AWS_ACCESS_KEY_ID	A valid AWS access key ID for the S3 bucket
AWS_SECRET_ACCESS_KEY	A valid AWS secret access key for the S3 bucket

Volume labels

Volume labels can overwrite the defaults from environment variables:

Label	Description
duplyvolume.remove_older_than	See REMOVE_OLDER_THAN
duplyvolume.remove_all_but_n_full	See REMOVE_ALL_BUT_N_FULL
duplyvolume.remove_all_inc_of_but_n_full	See REMOVE_ALL_INC_OF_BUT_N_FULL