Simple Dockerized S3 Backup Tool

Simple S3 Backup Synchronizer. Manages local backup folder with configurable maximum number of backup files and uses awscli (aws s3 sync) to synchronize the local backup with configurable Amason S3 location.

To show script help:

docker run --rm procraft/s3-backuper --help

Usage

Before you can use the backup script you should create your own derived docker image with your Amazon credentials:

mkdir -p path/to/some/dir
cd path/to/some/dir
wget https://raw.githubusercontent.com/procraft/s3-backuper-docker/master/configured-image/Dockerfile
docker build -t foo/s3-backuper --build-arg aws_id='your-aws-id' --build-arg aws_secret='your-aws-secret' --build-arg aws_region='us-west-2' .

You can replace foo/s3-backuper with any valid docker image tag. This image is created locally on your docker host and should never be shared because it contains your credentials.

Now you can call the backuper script. Let's assume you have a running docker container named my-service which periodically creates backup files in volume /data/backups/. You can create a cron job with command:

docker run --rm --volumes-from my-service foo/s3-backuper /data/backups/ s3.bucket.name s3/folder/my-service --keep 5

This will:

• Scan directory /data/backups for any new files and move them to the _processed subfolder with added date-prefix.
• Remove oldest files in _processed subfolder if there are more then 5 files (--keep option, default is 10).
• Create bucket s3.bucket.name (if it doesn't exist) in the configured previously region.
• Sync _processed/ subfolder with s3://s3.bucket.name/s3/folder/my-service/ using aws s3 sync with --delete switch.

License

MIT

Thanks

Plumbum