Amazon Web Services (AWS) is one of the most popular cloud providers out there. It consists of over 15 regions geographically located across 4 continents, and provides over 70 different services. It serves customers in over 190 countries, including several governments. Elastic Compute Cloud, better known as EC2, launched in 2006, and is one of the most popular services of AWS. Some of the most popular components of EC2 are:

• Instances (virtual machines)
• Volumes (disks)
• Amazon Machine Image or AMI (disk snapshots)
• Security groups (firewalls)
• Elastic Load Balancers (application and network load balancers)
• Auto scaling groups (to scale instances)

Most of the basic building blocks for running a simple application are provided by AWS EC2 itself.

Ansible interacts with AWS APIs to manage various infrastructure components. Using APIs requires credentials and an IAM user with the right permissions would help us do this. We also need to set up some libraries that would be required to execute certain Ansible modules. So let us get started.

Ansible ships with scores of AWS modules. These Ansible modules use AWS Python SDK, called Boto, as dependency and interact with AWS.

1. Let us install Boto using Python pip to get started:

$ pip install boto

2. Along with Boto, we also need to have a user who has enough privileges to create and delete AWS resources. AWS has a predefined policy called AmazonEC2FullAccess which can be...

Virtual Private Cloud, or VPC, is technically not a part of EC2. However, this is usually the first step when getting started with EC2. VPC creates a virtual network which logically isolates our resources. This improves security and management since, logically, subnet and gateway are dedicated for our resources only. A common usage of VPC is to isolate public-facing services (like load balancers or instances running public services) and servers storing data (like databases) which do not require direct access from the wider internet.

Technically, a VPC has several moving parts, as depicted in the preceding image. Even a simple architecture would consist of the following components:

• The VPC itself, where we will allocate a high-level Classless InterDomain Routing (CIDR) block and choose a region.
• A public subnet, which will use a chunk of CIDR from...

EC2 security groups are virtual firewalls, which control inbound and outbound traffic to and from our EC2 Instance. We will create security groups before an EC2 Instance because this resource is required for creating an EC2 instance. Security groups and EC2 instances have many-to-many relationships. We can have a single instance with multiple security groups and a single security group can be applied to multiple instances, even multiple AWS instances present in the same subnet can have different security groups.

We can create a security group, using an ec2_group module, this will take the VPC ID, the region, and rules as input.
Let's create a task for a security group...

Elastic Cloud Compute or EC2, forms a central part of Amazon Web Services. It is one of the most popular services by AWS, which provides rented virtual computers (called instances) under various capacities in terms of CPU, memory, disk, network, and so on where users can run and host their applications.

Before going ahead and launching an EC2 Instance, we will require resources we created in preceding tasks; that is, security groups and VPC (subnets).

Creating an EC2 instance requires the following basic parameters:

• Instance type, which determines hardware of the host computer used for our instance. An instance type offers a wide range of compute, memory, network, and storage capacity...

An Elastic IP address, also known as EIP is a static IPv4 address, which can be assigned to an EC2 instance. The Elastic IP address is generally used for covering the failure of an EC2 Instance by quickly remapping it to another EC2 instance. An Elastic IP address is allocated to an AWS account and can only be used in a specific region. That is, we cannot associate an Elastic IP address allocated in one region to an EC2 instance in a different region.

We can allocate and associate an Elastic IP address with an EC2 instance using an ec2_ip module. This will require the instance ID of the EC2 instance we want to associate this Elastic IP address with and the region of that instance...

Amazon Elastic Block Storage, also known as EBS, is block-level storage that can be attached to an EC2 Instance. Once we attach an EBS volume, we can create a file system on top of these volumes, mount it, and use it like an attached disk, technically as block storage. Amazon EBS volumes are placed in a specific availability zone and they can only be attached to an EC2 instance present in the same availability zone. EBS volumes are automatically replicated to protect our data from the failure of a single component.
In this section, we will be writing a task to create an EBS volume and attaching it to an existing EC2 instance (which was created in previous tasks).

We have already discussed Amazon Machine Image (AMI) while creating EC2 instances. AMI is the most common way to create backup images of an EC2 Instance. For the sake of consistency, while creating an AMI, AWS reboots the instance by default. However, this can be overridden and AMIs can be created without rebooting the EC2 Instance. AMIs store references to device mapping with corresponding volume snapshots. AMIs can be used to define a standard template which can be used while launching EC2 instances. For example, we may want to use a standard flavor of Linux with predefined system configurations across all EC2 instances. We can create an AMI for this and use it while creating EC2 instances.

Often, for the purpose of load balancing, as well as to achieve high availability, we may choose to serve the traffic using a load balancer. AWS provides a virtual load balancer, called Elastic Load Balancer (ELB), which can be used to receive traffic from clients and route the traffic to a set of instances which are attached to the ELB. Apart from stability, the ELB has quite a few features which can help in improving the overall uptime. One of the most important of these features is health check. This lets ELB determine that an attached instance has gone bad and it should stop routing traffic to it. ELB can also insert cookies which can be used for making routing decisions and it can be used to offload SSL from an application.

• Launch Configurations act a template for auto scaling groups to launch EC2 instances. It contains information like the AMI ID, the security group, and so on required to launch an EC2 instance.
• Auto Scaling Groups is a collection of EC2 instances which...

Our phonebook application can be deployed to the instances that we have created. When deploying an application to an instance, we either need to know the IP address of the instance and prepare the inventory, or we can figure out the IP address at runtime. Preparing the inventory is often simple, however, it requires manual intervention. We have to run tasks to boot an EC2 instance with the required parameters and copy the IP address of the instance to the inventory file. After this, we can run the playbook for deploying the application.

Manually adding IPs to the inventory is not possible for unattended setups. In certain cases, the infrastructure is dynamic to the extent that managing IPs might not even be possible. For such cases, there are two possibilities: we can use Ansible's add_host module to deploy an application when we boot up...