Glacier auto upload
The main purpose of this script is to automatically upload archives to Amazon Glacier from a selected folder on your machine. It works great on Raspberry Pi!
What does it do
The script tries to chunk all files within archive directory into manageable 50MB parts and sends them to your Amazon Glacier account, creating vaults based on folder names and uploading files to these vaults. It keeps track of files already uploaded as well as created vaults. If file is kept in a subfolder, the vault name will reflect that:
photos/holiday 2013/week-1.zip => photos-holiday/week-1.zip
photos/holiday 2013/week-2.zip => photos-holiday/week-2.zip
photos/dinner with friends/photos.zip => photos-dinner-with-friends/photos.zip
Requirements
- Linux machine that can stay on for days (or even months, depends on your upload speed and amount of data)
- installed and configured glacier-cmd
- installed 7zip
- Amazon AWS account
Installation
- Put the file in a convenient location, possibly in your home folder or in /usr/local/bin/
- Make sure that the file is executable:
sudo chmod +x glacier-auto-upload.sh - Point
ARCHIVE_DIRvariable to the directory containing your data - Add script to your crontab, executing as
bash glacier-auto-upload.sh
Known issues
- Can't upload files directly within the
ARCHIVE_DIR, they have to be kept in a subfolder within that folder - Failed uploads are not terminated automatically