Create ECS Cluster and deploy Rails + Sidekiq tasks

This article is the continuation of Rails + Sidekiq + Docker Application for AWS (ECS, ECR, RDS, Codepipeline and more) complete series. For better understating, please start from the beginning of the series.

In this article, our website will go live with Load Balancer and Autoscalling groups (having multiple Rails instances on demand). This means that AWS can increase the number of Rails instances on demand and it can scale it down too.

We will do the following in this article:

Create ECS Cluster
Create Service for Rails application and Sidekiq
Configure Autoscalling

Create ECS Cluster

Go to ECS from your AWS console service tab and select Clusters, Click on Create Cluster

Step 1: Select Cluster template

Select EC2 Linux + Networking

Step 2: Configure cluster

We have selected t2.medium for our instance type. Feel free to change it as per your requirement. Number of instances are set to 3 initially.

In the networking section, we will select the correct VPC the rails-app-instance-sg (The Security group for our rails instance created earlier)

Now it will take couple of minutes to create the cluster and the EC2 instances for you.

If you will go back to your cluster and click on EC2 instances tab, you will be able to see 3 instances (it may take AWS to some time to create the instances)

Create Rails service

We will now add a service to our container which will have the rails-application-task in it.

Step: Configure Service

We are setting Minimum healthy percent to 33, the reason is that we have 3 instances in the cluster. We’re simply saying keep at-least one task running while deploying. I will discuss this in more detail when we’ll do CI/CD using AWS Codepipeline

Click on Add to load balancer button select 80:HTTP and don’t forget to select the target group.

Set Auto Scaling group

We are about to create an Auto Scaling group policy that simply says that our desired tasks count is 3 and maximum is 4 if load increases.

Creating Scale UP and Scale Down policies.

Something has to trigger an alarm so AWS EC2 can add or reduce number of instances for us. I am using CPU Utilization in our example. There are many more options.

To keep it simple, we are saying that if the CPU Utilization is above 60% for 5 minutes, add another instance.

If the CPU Utilization is below 50% for 5 minutes, reduce instances to just 1.