Automating Amazon S3 backups on a Windows Server

1: Create an Amazon AWS account

If you don't already have an AWS account - create it here, it's free. Amazon's "free usage tier" on S3 gives you 5GB free storage from scratch, so after registering, sign in to your "AWS Management Console", select the "S3" tab and create one or more "buckets" on the left.

2: Get your access keys

You will need security credentials to access your online storage from the server, so click your account name - "Security Credentials" - "Access Keys" and copy your Key ID and Secret.

3: Download "S3Sync"

"S3Sync" is a great free command-line application from SprightlySoft. It is .NET-based and even comes with the source codes. At the time of writing this post their website was down, so I published the tool on Google Docs here: S3Sync.zip.

The tool syncs a given folder with your S3 bucket. And the best part - unlike similar scripts and utilities it performs a "smart" differential sync that detects additions, deletions and file-modifications.

4: Write a backup script

Create a batch file and paste this code into it:

S3Sync.exe -AWSAccessKeyId xxxxxxx -AWSSecretAccessKey xxxxxxx -SyncDirection upload -LocalFolderPath "C:\inetpub\wwwroot" -BucketName YOURBUCKETNAME

The code above is pretty self-explanatory. Just replace the "xxxxxx" with your access codes from #2, "YOURBUCKETNAME" with the name of your S3 bucket, and "C:\inetpub\wwwroot" - with the folder you want to backup. Then create a scheduled task that runs the batch file every 24 hours, and you're all set.