Manage Amazon S3 with s3cmd

Productivity Sauce

While Amazon S3 provides reliable and cheap backup, you need a third-party application to put the service to some practical use. JungleDisk is one of the most popular Amazon S3-based backup solutions out there, but it's not the only fish in the sea. If you are looking for a no-frills tool which can help you to manage the Amazon S3 service, try s3cmd -- a simple command-line utility written in Python.

Before you start, you have to sign up for the Amazon S3 service, if you haven't already done that. To install s3cmd on your machine, download the latest release of the utility, unpack the downloaded archive, use the terminal to switch to the resulting directory and run the python setup.py install command as root. Next, you have to configure it using the s3cmd --configure command (have your Amazon identification keys ready to enter when prompted). Once s3cmd is configured and ready to go, you have to create a so-called bucket -- a repository on Amazon S3 servers where you store your files. To do this, use the s3cmd mb s3://BUCKET command, where BUCKET is a unique name for your bucket. Keep in mind that the bucket name must be unique among all buckets on the Amazon S3 servers, so you have to come up with something more original than "myfiles" or "documents". Fortunately, you don't have to remember the exact name of the bucket, as you can use the s3cmd ls command to view all your existing buckets on Amazon S3. To upload a single file, use the s3cmd put command followed by the path to the file and the target bucket, for example:

s3cmd put /path/to/loremipsum.odt s3://20081113-1258zrtm7

To download a file from the bucket on your machine, you can use the s3cmd get command and specify the bucket and the name of the downloaded file:

s3cmd get s3://20081113-1258zrtm7/loremipsum.odt loremipsum.odt

If you want to back up an entire directory, uploading files one-by-one is not very practical. Fortunately, s3cmd features the sync command that allows you to synchronize a local directory with a directory in an S3 bucket:

s3cmd sync /path/to/dir s3://20081113-1258zrtm7/backup

Using the same command, you can also do a reverse synchronization which effectively acts as a restore operation:

s3cmd sync s3://20081113-1258zrtm7/backup /path/to/dir

To view a list of other commands supported by s3cmd, run the s3cmd --help command.