Michael Friis

Docker for AWS Public Beta

Michael Friis

Today, we’re announcing that Docker for AWS is graduating to public beta, just in time for AWS re:Invent.

Docker for AWS

Docker for AWS is a great way for ops to setup and maintain secure and scalable Docker deployments on AWS. With Docker for AWS, IT ops teams can:

  • Deploy a standard Docker platform to ensure teams can seamlessly move apps from developer laptops to Dockerized staging and production environments, without risk of incompatibilities or lock-in.
  • Integrate deeply with underlying infrastructure to ensure Docker takes advantage of the host environment’s native capabilities and exposes a familiar interface to administrators.
  • Deploy the platform to all the places where you want to run Dockerized apps, simply and efficiently
  • Make sure the latest and greatest Docker versions are available for the hardware, OSs, and infrastructure you love, and provide solid upgrade paths from one Docker version to the next.

To try the latest Docker for AWS beta based on the latest Docker Engine betas, click the button below:

Docker for AWS

Docker for AWS works fully within AWS free tier, giving you the ability to try it out at no cost (just create a 1-manager, 1-worker swarm). Installation takes a few minutes, and will give you a fully functioning swarm, ready to deploy and scale Dockerized apps.

We first unveiled the Docker for AWS private beta on stage at DockerCon 2016 back in June, and we are excited to be opening up to beta to the public. We received lots of great feedback from private beta testers (thanks!) and incorporated as much of it as possible. Enhancements added during the private beta include:

  • All container logs are automatically sent to AWS CloudWatch for later inspection and analysis. That means you no longer have to rummage around on hosts to find the error you’re looking for or worry that logs are lost if a worker is replaced.
  • Built-in diagnose tool lets you submit a swarm-wide diagnostic dump to Docker so that we can help diagnose and troubleshoot a misbehaving Docker for AWS swarm.
  • Configurable ephemeral instance root disk size and type lets you choose faster SSDs and bigger disks to hold all your images, containers and volumes.
  • Improved upgrade stability so that you can confidently upgrade your Docker for AWS to the latest version

We’re particularly proud of the progress we’ve made on diagnostics and upgradability. These are features that set a true production system apart from simple fire-and-forget templates that just spin up resources without thought for debugging or future upgrades.

The improvements added during the private beta complement the initial features Docker for AWS launched with earlier this year:

  • Simple access and management using EC2 keypairs
  • Quick and secure deployment of websites thanks to auto-provisioned and auto-configured load balancers
  • Secure and easy-to-manage EC2 network and instance configuration

With today’s public beta announcement, we hope to get even more users interested in running Docker on AWS and testing the beta. Check out the detailed docs and sign up on beta.docker.com to be notified of updates and new beta versions.

And if you’re looking for the next step in enterprise manageability, you can also evaluate Docker Datacenter in one click using Docker for AWS. This template benefits from all the improvements of Docker for AWS while adding enterprise authentication, role-based access controls, a web-based graphical interface, and end to end security, all in a highly-available deployment for both Universal Control Plane and Trusted Registry. Find out more about this template here or get a 30 day evaluation license from the Docker Store.

If you have questions or feedback, send an email or post to the Docker for AWS or the Docker for Azure forums.

Additional Resources


2 thoughts on “Docker for AWS Public Beta

  1. Is there documentation on how this is assembled? I see the docker4x images but no github. I want to use the images but craft my own SQS, Dynamo etc… The reason for this is security requirements mean I have to deploy into existing VPC and subnets and tighten the IAM quite a bit to just the absolute ports/permission required for the swarm to function. I'd also want to deploy multiple swarms per VPC which would be a bit different. I can and have reverse engineered the CFT but I have no idea what kind of api calls are happening inside those instances. (Not sure if I can glean everything from userdata) So docs / source would help tremendously.

Leave a Reply