Motivation for backup

Since there is important information on the mediaserver from a historical point of view (i.e.: pictures from previous years) a backup is implemented to make sure that this data is backed up.

How the backup is implemented

The backup of the pictures is implemented by means of a docker image of duplicati. Duplicati is an open-source backup tool that can be accessed through a web interface on the local network through port 8200.

URL: 10.0.4.33:8200

Duplicati then backs up the data to a 1TB remote drive at stackstorage.com with webaddress sgeboers.stackstorage.com, a seperate account for ds4 with password roodblauw222 was added such that ds4 can also access the drive through the webdav protocol. This is all entered in the specific backup Photobackup within the duplicati web-interface.

A few notes:

  • Duplicati does not simply copy the files, instead it hashes them and divides/merges the files into blocks of 50 MB. Apart from this data file a json file is created with the data files contents. These files have the advantage that bandwidth and storage space is saved as well as giving a good foundation for possible needed restorations later on. Since it does this all, it will take quite a while for the initial backup to finish.

  • During the creation of the docker image it was linked to three logical docker volumes: a config volume, a backup volume and a source volume. In case you want to back up something else; you need the source volume. In the web interface you will see a virtual environment of the docker image (ie it will not look like the file system of the mediaserver). However, if you go to the map source (which is linked to the root of the mediaserver, that is "/:/source") you can find other maps of the mediaserver which you might want to backup (for example install files).