Home

Awesome

python version Code style: black Travis (.com) PyPI Stackshare: runrestic PyPI - Downloads

Runrestic

runrestic is a simple Python wrapper script for the Restic backup software that initiates a backup, prunes any old backups according to a retention policy, and validates backups for consistency. The script supports specifying your settings in a declarative configuration file rather than having to put them all on the command-line, and handles common errors.

Example config

repositories = [
    "/tmp/restic-repo",
    "sftp:user@host:/srv/restic-repo",
    "s3:s3.amazonaws.com/bucket_name"
    ]

[environment]
RESTIC_PASSWORD = "CHANGEME"

[backup]
sources = [
    "/home",
    "/var"
    ]

[prune]
keep-last =  3
keep-hourly =  5

Alternatively you can also just use JSON. For a more comprehensive example see the example.toml or check the schema.json

Getting started

Installing runrestic and restic

To install runrestic, run the following command to download and install it:

sudo pip3 install --upgrade runrestic
<br> You can either manually download and install [Restic](https://restic.net/) or you can just run `runrestic` and it'll try to download it for you.

Initializing and running

Once you have restic and runrestic ready, you should put a config file in on of the scanned locations, namely:

Afterwards, run

runrestic init # to initialize all the repos in `repositories`

runrestic  # without actions will do: runrestic backup prune check
# or
runrestic [action]
<br> Certain `restic` flags like `--dry-run/-n` are built into `runrestic` as well and will be passed to restic where applicable.

If, however, you need to pass along arbitrary other flags you can now add them to the end of your runrestic call like so:

runrestic backup -- --one-file-system

Logs for restic and hooks

The output of restic and the configured pre/post-hooks is added to the runrestic logs at the level defined in [execution] proc_log_level (default: DEBUG), which can be overwritten with the CLI option -p/--proc-log-level.

For process log levels greater than INFO the output of file names is suppressed and for log levels greater than WARNING restic is executed with the --quiet option. If the process log level is set to DEBUG, then restic is executed with the --verbose option.

It is also possible to add restic progress messages to the logs by using the CLI option --show-progress INTERVAL where the INTERVAL is the number of seconds between the progress messages.

Restic shell

To use the options defined in runrestic with restic (e.g. for a backup restore), you can use the shell action:

runrestic shell

If you are using multiple repositories or configurations, you can select one now.

Prometheus / Grafana metrics

@d-matt created a nice dashboard for Grafana here: https://grafana.com/grafana/dashboards/11064/revisions

systemd timer or cron

If you want to run runrestic automatically, say once a day, the you can configure a job runner to invoke it periodically.

systemd

If you're using systemd instead of cron to run jobs, download the sample systemd service file and the sample systemd timer file. Then, from the directory where you downloaded them:

sudo mv runrestic.service runrestic.timer /etc/systemd/system/
sudo systemctl enable runrestic.timer
sudo systemctl start runrestic.timer

cron

If you're using cron, download the sample cron file. Then, from the directory where you downloaded it:

sudo mv runrestic /etc/cron.d/runrestic
sudo chmod +x /etc/cron.d/runrestic

Changelog

Ansible

@tabic wrote an ansible role, you can find it here: https://github.com/outwire/ansible-role-restic . (I have neither checked nor tested it.)

Development

This project is managed with poetry

Install it if not already present:

curl -sSL https://raw.githubusercontent.com/python-poetry/poetry/master/get-poetry.py | python
# or
pip install --user poetry

Installing dependencies

poetry install

Running Tests

poetry run pytest

Using VScode devcontainer

The project contains a .devcontainer folder with the settings for VScode to develop inside container. The Python virtual environment created by poetry is stored outside the container in the projects path .virtualenvs so that it survives container rebuilds.

The Ubuntu 22.04 based container uses Python 3.10 as system version and includes minimal Python 3.7, 3.8 and 3.9 versions for creating virtual environments in any of those versions.

It is possible to switch the Python version used by poetry with the command poetry use <version>, see poetry managing environments for more details.

Thanks

This project was initially based on borgmatic but has since evolved into something else.