Backup selected directories to S3

Script reads file with directories list to backup and copy them to S3 as archives or direct copy (sync). The script is a compilation of different examples and approaches I found in internet which I adapted for my needs.

Prerequisites

• s3cmd is installed and properly configured
• create s3backups dir in /tmp with 700 permissions
• edit dirs.txt with actual paths - each path per line.

Examples

Minimal example:

bash dir-backup-s3.sh --m sync --s s://path-to-bucket/dir-example/

Complete example:

$ bash dir-backup-s3.sh \
    --mode archive \
    --webhookurl https://path-to-webhook.example \
    --range 1_week_ago \
    --s3url s://path-to-bucket/dir-example/ \
    --name MySuperServer \
    --dirs dirs.txt \

Arguments

Short name	Verbose name	Description
--m	--mode	sync or archive. Use sync to direct copy to S3 (good for heavy content), archive - make archives first then upload to S3.
--w	--webhookurl	Path to webhook URL. Useful for external integrations.
--r	--range	Use format like: 2_days_ago,1_week_ago etc. How many backups are going to be made. Underscore as space.
--s	--s3url	Path to S3. Starts with s3:// and ends with trailing /.
--n	--name	Name of the server. Hostname is used by default.
--d	--dirs	Path to file with directories list to backup (each path per line).