Skip to content

Latest commit

 

History

History
128 lines (105 loc) · 5.76 KB

Terraform_Resources.md

File metadata and controls

128 lines (105 loc) · 5.76 KB
Raw

Terraform Resources

Component or Object

Terraform has two different type of component/object

  • resource
  • data

Resource:

Resource are the daily bread of Terraform. They illustrate the infrastructure pieces that you want to manage such as networks, servers, firewalls, etc. Terraform will use the cloud provider APIs to perform the create, read, update, and delete(CRUD) operations. The resource object is constructed of a provider-name_resource-type, local identifier and the block containing the configuration of the resource. This would be better understood with the below diagram.

All the resources are linked to a provider. From the above diagram, aws_instance indicates that this resource type is provided by the aws provider. Next is the local identifier name for the resource, which is specified by you, here we have named as web so that we can reference it elsewhere and Terraform keep track of it in the .tfstate file. Concept of state file will be covered in the upcoming tracks.

See the below example code for the resource creation and how it is referenced to another piece of resource block.

resource "aws_instance" "collabnix_node" {
 ami               = "ami-21f78e11"
 availability_zone = var.availability_zone
 instance_type     = var.instance_type

 tags {
   Name = "Collabnix_terraform"
 }
}

resource "aws_ebs_volume" "collabnix_vol" {
 availability_zone = "us-east-1c"
 size              = 1

  tags {
   Name = "Collabnix_terraform_vol"
 }
}

resource "aws_volume_attachment" "collabnix_vol_attachment" {
 device_name = "/dev/sdh"
 volume_id   = aws_ebs_volume.collabnix_vol.id
 instance_id = aws_instance.collabnix_node.id
}

From the above, we are creating a single EC2 instance, EBS volume, and attaching that volume to the instance. The first two resource blocks (EC2 instance & EBS volume) will be created independently. While trying to attach the volume to the instance, Terraform requires IDs of instance and volume to be attached. In this case you need to refer the local identifier name and the required attributes of the resources. See the below diagram for better understanding:

Now terraform has enough information to take the necessary action. Here the id attributes are accessed using the dot-separated notation, like aws_instance.collabnix_node.id. Each type of resource will export thier own set of attribute values. Here, for example we used the id attribute. Refer this link for more details (Attribute reference)

Data sources:

These are very similar to regular resource object which represents a piece of read-only information that can be fetched from the provider (here it is AWS) or from an external data source. This cannot be used for any operations like CREATE, UPDATE, or DELETE. Instead, they can only return several informations (meta-data) like AMI ID, Private IP, Instance ID and so on from an existing resources.

data "aws_ami" "app-ami" { 
  most_recent = true 
  owners = ["self"] 
} 
 
resource "aws_instance" "community" {
  ami           = data.aws_ami.app-ami.id
  instance_type = "t2.micro"
  tags = {
    Name = "Collabnix"
  }
}

From the above example code, we are creating an EC2 instance by using the existing AMI ID. Let's assume we have already created an AMI manually or with a different set of tools like Packer. Now terraform needs AMI-ID to create an instance and it fetches the ID from the data source app-ami.

Note: The combination of resource type and the local identifier name must be unique in your configuration. The below configuration will through an error like: aws_instance.collabnix_node: resource repeated multiple times

resource "aws_instance" "collabnix_node" {
 ami               = "ami-21f78e11"
 instance_type     = var.instance_type
}

resource "aws_instance" "collabnix_node" {
 ami               = "ami-21f78e11"
 instance_type     = var.instance_type
}

Arguments

This is just a syntax of assining the vaules within the configuration blocks. It looks like the below

resource "aws_instance" "collabnix_node" {
 ami               = "ami-21f78e11"          # <IDENTIFIER> = <EXPRESSION>
 instance_type     = var.instance_type
}

Each type of resources will have the list of supported arguments (required and optional) you can consume within your configuration blocks.

Morover, you can also use the special kind of arguments called meta-arguments within any type of resource. Primarily, these meta-arguments are used to change the behavior of the resource. See the list of meta-arguments below

  • depends_on
  • count
  • for_each
  • provider
  • lifecycle
  • provisioner
  • connection
resource "aws_instance" "collabnix_node" {
  depends_on        = [aws_s3_bucket.collabnix_bucket]
  count             = 2
  ami               = "ami-21f78e11"
  instance_type     = var.instance_type
}

resource "aws_s3_bucket" "collabnix_bucket" {
  bucket = "lab-bucket"
  acl = "private"
  versioning {
    enabled = true
  }
  tags {
    Name = "test-s3-terraform-bucket"
  }
}

The above configuration blocks are just an example of how the meta-arguments can be used. Please do not go deeper and try to understand how it works in this beginner track.