serac 0.0.3

Incremental permanent data archiver with encryption

Incremental permanent data archiver with encryption.

Designed for permanently backing up data which does not change frequently, suitable for write-only storage such as Amazon Glacier.

Installation

This requires Python 3.7.

Install serac with:

pip install serac

If you don’t have Python 3.7 installed, you can install it with pyenv:

curl https://pyenv.run | bash
# restart your shell
pyenv update
pyenv install 3.7.3
pip install virtualenv
virtualenv python=python3.7 venv
. venv/bin/activate
pip install serac

Usage

Serac must always be run with the config file path as the first argument, then the command to perform as the second argument.

To run serac:

/path/to/venv/bin/serac CONFIG COMMAND [OPTIONS]

It is safe to run Serac from a cron job; it will not allow multiple processes to work with the same config file at the same time.

Commands

After that it accepts one of the following commands:

test

Test the configuration file

init

Initialise an index for a new config by creating the database

archive

Archive any changes since the last archive was performed

ls [--at=DATE] [--pattern=PATTERN]

Show the state of the archive.

This follows the roughly the same layout as ls -l, with the following columns:

• File permissions

• Owner (as it will be restored to on this system)

• Group (as it will be restored to on this system)

• Size (in kibi/mebib/gibibytes, or in bytes if not specified)

• Last modified date (this year if not specified)

• Last modified timestamp (for ease of use in calls to ls and restore)

• Path (as it was on the originating system)

restore DESTINATION [--at=DATE] [--pattern=PATTERN]

Restore some or all of an archive

If an archived object is in glacier it will be marked for retrieval.

Arguments

DATE

This should be a date in one of the following formats:

• epoch timestamp, eg 1582165202

• YYYY-MM-DD, eg 2020-02-20

• YYYY-MM-DD HH:MM:SS, eg 2020-03-20 02:20:02

• YYYY-MM-DDTHH:MM:SS, eg 2020-03-20T02:20:02

PATTERN

This can either be an exact path to a file, or a partial path to a directory.

Globs are not yet supported.

Configuration

Configure serac using a config file:

[source]
# Define the source for the backups

# List of paths to include and exclude (glob patterns)
include =
    /path/to/source
    /path/somewhere/else
exclude =
    /path/to/source/unprocessed
    /path/somewhere/else/*.jpg

[archive]
# Define where the backups are saved

# Backup to a local path
#storage = local
#path = /path/to/backup

# Backup to S3
storage = s3
key = 4p1_k3y
secret = 53cr3t
bucket = arn:aws:s3:::my_bucket_name
path = path/within/bucket

# Encrypt backups with this password
password = l0ng_s3cr3t

[index]
# Define how indexed files are treated

# Location for index database
# This should then be backed up by another service, eg duplicity
path = /path/to/index.sqlite

Contributing

To work on serac, install it in a virtual environment:

mkdir serac
cd serac
git clone <path-to-repo> repo
virtualenv --python=python3.7 venv
. venv/bin/activate
pip install pip-tools
cd repo
pip-sync

To run during development:

python -m serac CONFIG COMMAND [OPTIONS]

To run tests:

cd serac/repo
. ../venv/bin/activate
pytest

Changelog

0.0.3, 2019-10-12

Bugfix:

• Correct entrypoint

0.0.2, 2019-09-25

Feature:

• Add process locking

0.0.1, 2019-09-23

Feature:

• Initial release