Skip to content


Subversion checkout URL

You can clone with
Download ZIP
Simple backup tool with pluggable storage backends (S3, FTP, FS) and encryption.
Branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.

UmaticsSync - Backup as customizable as you want

UmaticsSync helps you to backup whatever and whenever you want. It provides pluggable storage backends. Currently it supports:

  • S3
  • FTP
  • Filesystem

It also makes use of GnuPG for encrypting the backups before storing them. We use it in production only with the S3 storage backend. Others are untested.


You need the following dependencies:

  • Python > 2.4
  • Python sqlite3
  • Python boto
  • SQLite 3 executables
  • GnuPG executables

First install the binary dependencies (here for Gentoo):

USE="sqlite" emerge python sqlite3 setuptools gnupg

Next get the Python dependencies:

easy_install sqlite3 boto

Obtain the source code:

git clone

Edit the backup jobs in conf/backup_list (examples below) and copy them to ~/.umaticssync (create the directory first).

mkdir ~/.umaticssync && cp conf/backup_list ~/.umaticssync

Now you can start the backup script and deamonize it:

python >/dev/null 2>&1


screen -dmS backup python


This is a sample backup job, that shows the various options:

interval = 21600
post_command = rm /tmp/dump.tar && rm -Rfv /tmp/dump
destination = s3://
path = /tmp/dump.tar
pre_command = mongodump -d db -o /tmp/dump && tar -cpf /tmp/dump.tar /srv/dump
revisions = 10
encrypt =
compress = true

The value in the brackets is the name of the job. This name is later used as a directory name on the target storage device. Make sure this name has no fancy chars in it.

interval = <seconds>

Time between backups

post_command = <cmd>

Shell command that is run after the backup (e.g. cleanup)

destination = (s3|ftp|file)://<username>:<password>@<host> 

Please refer to the storage wrappers to find the details on the usage

path = <path-to-backup> 

Should be a file or directory

pre_command = <cmd> 

Will be run before the backup starts (e.g. compress files)

revisions = <number> 

How many revisions would you like to keep

encrypt = true|false 

Use gnupg to encrypt backup

compress = true|false 

Let umaticssync compress (using bzip2) the files for you


# Normal backup for a specific job
$ ./ -j <job>

# Restore job with name documents and revision number 32
$ ./ -j <job> -r 32

# Run in foreground and execute every job in the given interval
$ ./


UmaticsSync and its creator are not responsible for any damages, or missing files. Use this at your own risk.

Something went wrong with that request. Please try again.