Manual Installation - Terraform Configuration

Create S3 Bucket for AWS Cloud

Warning

Use only the AWS CLI for bucket creation. Do not create a bucket through the AWS Management Console.

1. Use AWS CLI to create a new S3 bucket.

2. Note the bucket name, region, and key.

Create a Directory and Set Up Terraform Configuration

1. In a terminal, navigate to the environment’s main directory.

2. Create a new directory with the S3 bucket name:

   mkdir -p buckets/environments/<bucket_name>
   

3. Copy the backend.tf and variable.tfvar files from the example directory into the new bucket directory:

   cp example-config/buckets/{backend.tf,variable.tfvar} buckets/environments/<bucket_name>
   

   The backend.tf file specifies where to store the Terraform state for this bucket deployment. Store the Terraform state with the backend.tf and variable.tfvar file here. See the following example:

   path="environments/<bucket_name>/terraform.tfstate"
   

4. Edit the variable.tfvar file.

5. Change the path, region and bucket to match the AWS S3 bucket settings. See the following example:

   variable.tfvar example

   path="path_to_tfstate"
   region="us-west-2"
   bucket="<bucket_name>"
   

6. Apply the Terraform configuration with the bucket name. This applies the configuration for the buckets module with the <bucket_name> from the backend.tf and variable.tfvar:

   make apply module=buckets env=<bucket_name>
   

7. Type yes to complete the configuration.

Scale-Related Terraform Configurations

This section summarizes all scale-related Terraform configurations in the variable.tfvar file. You will need to adjust the following according to the target deployment scale:

// The spec of each K8s node
// Recommendations: Start with t3.2xlarge. Increase for higher scale as per platform system requirements section.
variable "eks_node_instance_type" {
   default = "t3.2xlarge"
}

// Each instance type can support up to a certain number of pods per node.
// Find the right eks_max_pods value for a given instance type at
// https://github.com/aws/amazon-vpc-cni-k8s/blob/master/misc/eni-max-pods.txt
variable "eks_max_pods" {
   default = 58
}

// This determines the Aurora Capacity Units (ACU) for the database
// Recommendations: Start with 2 max ACU. Increase for higher scale as per platform system requirements section.
variable "aurora_max_acus" {
   type        = number
   default     = 2
}

// Need to switch to "elastic" mode for high throughput
// Recommendations: Always use "elastic"
variable "efs_throughput_mode" {
   type    = string
   default = "bursting"
}