ResticBackupAutomation

The goal of the project is to provide a simplistic and reliable automated backups as well as some bash aliases to make manual backups a breeze. In case of this project remote backups are aimed at B2 cloud storage. The main difference between manual and automated backups is that manual tag is added to manual backups, unlike automated ones only having OS or Data tag. Such approach allows for additional control when deciding which backup snapshots to keep and which to remove.

The key concept of the ResticBackupAutomation is segregation between system data and user data.

• System data (OS tag) includes data starting from system root (/), /home too, except for the specified home user data directories (Pictures, Documents, Music, etc.).
• User data (Data tag) is focused on most of data in /home, but doesn't include /home/*/.* (any files/directories in user folder that start from . and belong to system/application configuration/data).

The reason for such distinction is because application configuration should be considered as a part of the OS and not the user files. Restoring the OS backup should restore, not only the system, but also the apps to the state and configuration they've been at for each user at the specific moment of time. Nevertheless, the majority of directories for mounted devices, temporary data and cache have been excluded from OS backups to preserve extra storage.

The project follows a user-centric approach that is it'll stay minimal, but each user is welcome to analyze and adapt everything to their needs.

In order to start using the automation it's necessary to:

• Ensure you're using GNU/Linux distribution with systemd as init.
• Copy all files except for .bashrc into the respective directories.
• Merge contents of .bashrc with the same file in your user directory, i.e. a couple of provided aliases could just be added to your file.
• Replace all placeholders (<>) in all files with real names/paths.
• Ensure that the following environment variables are defined in your /etc/environment file:
  • RESTIC_PASSWORD_FILE or RESTIC_PASSWORD
  • RESTIC_CACHE_DIR (usually is set to "/var/cache/restic")
  • B2_ACCOUNT_ID (application key id that has read/write access to your B2 bucket)
  • B2_ACCOUNT_KEY (application key that has read/write access to your B2 bucket)
• In case of RESTIC_PASSWORD_FILE environment variable, the file that contains restic repository password has to be created first.
• Run sudo systemctl daemon-reload
• Run sudo systemctl enable --now restic-local-backup.timer
• Run sudo systemctl enable --now restic-cloud-backup.timer
• Reboot.

The automated backups will run on a daily basis. restic-local-backup.timer will execute restic-local-backup.service, while restic-cloud-backup.timer will execute restic-cloud-backup.service. restic-cloud-backup.timer activates when a network is online. In addition restic-cloud-backup-service waits for completion of restic-local-backup.service to eliminate unnecessary CPU load, spikes and freezes when two restic processes are spawned and run in parallel.

Each of services calls the restic_backup bash script. The path to restic repository is provided as a parameter. The restic_backup script also performs repository pruning to clean up the deleted snapshots and free up disk space. This happens if it's executed on the 1st of month.

Manual backups could be done using sysbackup or databackup bash aliases. In addition recent snapshots could be accessed with busnapshots alias. All three commands first target a local restic repository, then a remote restic repository.

The restic itself is a minimal one-file command-line utility written in Go that is capable of encrypted, compressed, deduplicated and incremental file backups to various repositories, both local and remote. For more details on how restic works as well as ideas for customization of the backup approach to suit your needs, please refer to the restic official documentation.