Navigation

• Create a VPC
• Connect a new environment to a VPC
• Connect an existing environment to a VPC
• ELB Caveats

Back in 2009 AWS introduced VPC (Virtual Private Cloud) which allowed AWS resources to be provisioned in a more granular way. For example, you could choose the private IP address to associate with a particular instance through a network interface.

Engine Yard Cloud was, unfortunately, tightly coupled to the old way of doing things (EC2-Classic in AWS terminology). Sometime around 2011/2012, AWS started creating new accounts (or new regions) with a default VPC, and all but discontinued development of EC2-Classic at that time. They did, however, provide a compatiblity interface to allow existing workflows to remain intact while still allowing access to the niceties of VPC. Since that time, these “Default VPC” regions and the EC2 Classic compatibility layer were the only way we supported the concept of VPC.

Engine Yard has a large number of AWS accounts that are old enough to not have a default VPC. These customers have not been able to utilize new features, such as current generation instances, as there has been no support for custom VPCs in the Engine Yard platform up until now.

Utilizing VPC and ClassicLink, customers now have the ability to provision a VPC and add new instances to environments while maintaining all of their existing infrastructure. The rest of this post will outline how to get started with a VPC

Create a VPC

You’ll find a link to the Networks management page under the Tools dropdown on your dashboard

On that page, click Add Network

You’ll be greeted with the following form

• Select the appropriate account if you have more than 1
• Make sure the region matches the region of the environment(s) you plan to connect to the VPC.
• The CIDR will need to be in the 10.0.0.0/16, 10.0.1.0/16, 172.16.0.0/12, or 192.168.0.0/16 range so as to not clash with the EC2 Classic IP space. Note that your CIDR cannot be larger than a /16.
• Check the ClassicLink box if you intend to connect this VPC to an existing environment in the EC2 Classic space

When you click create, you’ll be taken back to the networks listing. Refresh this page after a few minutes, and you should see the new network listed. If you click on the VPC ID, you can see more information about the VPC you just provisioned

By default, we create a /24 for each availability zone that an account has access to. In this case, it happens to be all 5 availability zones in us-east-1.

Connect the VPC to an environment

Connect the VPC to a new environment

If you don’t want to utilize ClassicLink with an existing environment, go ahead and create a new environment. On the form, make sure to select the network from the dropdown

On the boot environment screen, you’ll have access to current generation instances. You should be all set at this point.

Connect the VPC to an existing environment

If you navigate back to your environment, and click edit, we can now select the network you just created. This field will only show you networks in your environment’s region.

As the field help says, once you click Update Environment, your instances will be connected to the VPC. A new firewall will be provisioned for your environment (leaving the old one intact for as long as there are ec2 classic servers in your environment). You’ll see your environment go through the following states

First the VPC ID will appear, the firewall will “disappear”, don’t worry it’s not really gone, and a new one will start provisioning.

Second, you will see the new firweall id

Third, the environment will revert to a normal looking state with the VPC ID showing permanently as a reminder that this environment is connected to a VPC.

On the Add instances page, you’ll see you now have access to current generation instances.

You’ll see the current generation instance booting along side the previous generation instance

Caveats

ELBs

Due to AWS limitations, a new ELB will be required if you opt to connect previous generation instances to a VPC using ClassicLink. An ELB must be created in a VPC in order to have VPC instances. However, an ELB created in a VPC can have ClassicLink instances connected to it.

You will need to create a new ELB, and choose to associate it to the environment so that the ELB is created with the environment’s VPC firewall.

Once the ELB finishes creating, your ClassicLink enabled instances will be part of the load balancer.

As you can see, the ELB is part of the VPC, and the instance connected is the solo instance from the screenshot above.

Once you finish creating this ELB, this would be a great time to change DNS. Once DNS propagation has finished (plan 24 hours or longer, depending on the TTL of the domain pointing to the ELB), you can start adding additional infrastructure inside the VPC. This way, you minimize costs of running extra servers during the time you are waiting for DNS propagation.

Press release

Austin (TX), USA (April 25, 2017) – Engine Yard, provider of the best cloud platform & support for Ruby on Rails, today announced that Crossover, the leader in cloud-based Ruby development teams, has acquired the company to develop the world’s best full-stack Ruby on Rails platform.

“Engine Yard has a decade of deep Ruby experience at the platform & support level and Crossover is the largest provider of Ruby developer cloud teams,” said Andy Tryba, CEO of Crossover & the new Engine Yard, and added: “By combining these two – we can now move up the stack and create a full stack Ruby on Rails experience to be a one-stop shop for the 1M+ Ruby applications out there.”

Read More

Many customers have received varying AWS instance notices related to instance retirement or instance maintenance. These retirement notices often read, “We have important news about your account. EC2 has detected degradation of the underlying hardware hosting your Amazon EC2 instance in the region.” Similarly maintenance notifications read, “One or more of your Amazon EC2 instances is scheduled for maintenance on for 2 hours starting at .” The big question is: do you know what these notices really mean and how you take care of them? Below we’ll discuss AWS notifications and how to act on them. The two biggest takeaways from theses notices are: run don’t walk, on working to replace affected instances; and make sure your mission critical data lives on Elastic Block Storage (EBS).

Read More

In the coming month we’ll be introducing Phoenix on Elixir as a first class framework and language on the Engine Yard Cloud Platform. Currently Engine Yard supports Ruby, PHP, and NodeJS. Elixir is a developer friendly, performant, and reliable language. Our offering will be releasing with the following features:

1. dashboard and cli based releases
2. dashboard and cli based database migrations
3. preconfigured cluster of Beam nodes.

Our roots started in the Ruby and Rails communities. We have since received multiple requests from our customers for our support of Elixir. We noticed a gap between deploying applications yourself and using an automated deployment platform that doesn’t allow full network access for OTP operations. Leveraging our managed service offerings we’ll be able to offer a fully configured OTP cluster on startup. This is because the instances in an environment we provision are within their own network. Stay tuned for more information as it becomes available from our engineering team as the language is brought into beta. If you would like to be included in our beta of managed Phoenix on Elixir please contact our customer success team.

Want to know more about Phoenix and Elixir? Check out the following resources:

• Phoenix Framework Documenation
• Elixir’s official Documenation
• Elixir for Rubyists

To effectively route traffic to your various instances Amazon’s Elastic Load Balancer(ELB) needs to know which of your instances are available. The configuration for how this is done is via the ELB healthcheck value. As the health check can directly impact your uptime you’ll want to select the right value, but what will that be for your specific case?

Read More