FakeS3 is a lightweight server that responds to the same calls Amazon S3 responds to. It is extremely useful for testing of S3 in a sandbox environment without actually making calls to Amazon, which not only require network, but also cost you precious dollars.
The goal of Fake S3 is to minimize runtime dependencies and be more of a development tool to test S3 calls in your code rather than a production server looking to duplicate S3 functionality. Trying RiakCS, ParkPlace/Boardwalk, or Ceph might be a place to start if that is your goal.
FakeS3 doesn't support all of the S3 command set, but the basic ones like put, get, list, copy, and make bucket are supported. More coming soon.
As of version 0.2.4 Fake S3 has a setting called '--cachedstore' that will cause it to read contents of the underlying file system from disk on every request. This is very helpful in cases where automation tasks need to reset the state of Fake S3 between runs.
gem install swirl-fakes3
To run a fakes3 server, you just specify a root and a port.
fakes3 -r /mnt/fakes3_root -p 4567
Take a look at the test cases to see client example usage. For now, FakeS3 is mainly tested with s3cmd, aws-s3 gem, and right_aws. There are plenty more libraries out there, and please do mention if other clients work or not.
There are some pre-requesites to actually being able to run the unit/integration tests
Edit your /etc/hosts and add the following line:
127.0.0.1 posttest.localhost
Then ensure that the following packages are installed (boto, s3cmd)
> pip install boto
> brew install s3cmd
Start the test server using
rake test_server
Then in another terminal window run
rake test
It is a still a TODO to get this to be just one command