ugration** under development **

Description

I have bunch of Linux boxes (work laptop, laptop and PC at home, 3 VPS etc) and want to have bacups, obviously.

Because of every box have own data to backup (~/.thunderbird on home PC, or /var/www/vhosts on my VPS, for example) - I made script to have all in one.

myBackup is flexible - thanks to its configuration file, where you can specify which resources (databases, files/dirs on filesystem), when (see runon option below, which allows to create backups by specifing weekdays) and where (see the store option - save to local filesystem, S3, Glasier) save backup to. More details about optionis - in the Configuration section.

It's under development when have some free time. Please - see ToDo features and ToDo development process lists at bottom.

Can backup:

local files;
#todo: remote files (via SSH);
#todo: local database (MySQL);
#todo: remote database (MySQL).

myBackup can save backups locally or remote - to the AWS S3 (#todo - AWS Glacier) bucket.

Workflow

Script uses myBackup's configuration file (conf/mybackup.ini by default) to get objects (an items, described in config's sections) to create backup from.

Each such an item must be added to the configugration file as dedictaed section, e.g.:

...
[KeyPass]
source = '/home/setevoy/Dropbox/KeyPass/'
destination = '/home/setevoy/Backups/KeyPass/'
store = 'local'
bkptype = 'full'
...

Here - full /home/setevoy/Dropbox/KeePass/ directory content will be saved in a KeePass_*type*_*date*_tar.gz file in the local directory /home/setevoy/Backups/KeePass/

Please note - section name (e.g. - "KeePass") will be used for archive name + bkptype (full or inc) and date (e.g. 2017_01_03_17_30_35). Result file will be KeyPass_full_2017_01_03_17_30_35.tar.gz.

Script will load defaults section first, and will overwrite each parameter with data from item's section.

You can use any scheduler (e.g. cron for Linux) to run script with desirable periods.

Prerequisites

Python modules:

ConfigParcer
argparse

Script tested on Python3/Linux.

Configuration

All configuration can be done via config file (conf/mybackup.ini by default).

You can specify another config file using -c option, e.g.:

$ ./mybackup.py -c anotherconfig.ini

Config file options list:

accesskey: AWS Access key. AWS_ACCESS_KEY_ID variable will be used if accesskey empty (default).
secretkey: AWS Secret key. AWS_SECRET_ACCESS_KEY variable will be used if secretkey empty (default).
datatype: file - for filesystem data (files, directories) or db - for databases (not implemented yet).
bkptype: full or inc (incremental) (see period for inc backup type).
source: for the file datatype and local store only - source file/directory to create tar.gz archive from.
destination: for file datatype only - destination directory to store created tar.gz file.
store: destination to store created tar.gz archives - local or s3.
period: period for the inceremntal bkptype to search changed files during last period (hours).
bucket: for the s3 store only - bucket name for backups.
droponpush: for the s3 store only - delete (1, "yes", "true", "True") or leave local files from the the desctination path after upload to S3.
fullretain: full backups number to store in store.
incretain: incremental backups number to store in store.
runon: days to run backup on item. Available options are: 'mo', 'tu', 'we', 'th', 'fr', 'sa', 'su'. Default - daily.

You can find examples in the existing config file.

ToDo features

options: overwrite section's parameters with command line options;
encrypt backups?

ToDo development process