This repository contains the configuration parameters for the Cloudprem infrastructure deployed via our CodePipeline configuration. Each directory represents one environment (e.g. development) and contains parameters such as the version of the infrastructure and the inputs for the Terraform stack.
This configuration repository is meant to be used as an upstream for your own config repo. To get started you can fork this repo to your own source control and make changes to the existing environments or add new ones. You can have 1 of each environment per region.
By default, when deploying our CodePipeline solution using the link above we pre-fill this repository's address as your
config repo but this is not recommended for production deployments as you will not have control over the configuration
or version of your infrastructure. It's much better to fork a copy and use that repository address instead during that setup
process. (See Module Config & Updating below for upgrade instructions)
This repository is broken down into the following sections:
- region
- environment
- module
- environment
At each level a configuration file exists that sets variables for that level.
As an example the us-east-1 region has a config file region.hcl inside the us-east-1 folder
that looks like this:
locals {
aws_region = "us-east-1"
}This aws_region variable should be edited to reflect the aws region if any new regions are added. Adding a new region is
as simple as creating a new region folder at the same directory level and naming it.
An example directory structure with two regions:
us-east-1/
├── region.hcl
├── development/
│ ├── env.hcl
│ ├── logical/
│ │ └── terragrunt.hcl
│ └── physical/
│ └── terragrunt.hcl
├── qa/
│ ├── env.hcl
│ ├── logical/
│ │ └── terragrunt.hcl
│ └── physical/
│ └── terragrunt.hcl
└── production/
├── env.hcl
├── logical/
│ └── terragrunt.hcl
└── physical/
└── terragrunt.hcl
us-west-2/
├── region.hcl
├── development/
│ ├── env.hcl
│ ├── logical/
│ │ └── terragrunt.hcl
│ └── physical/
│ └── terragrunt.hcl
├── qa/
│ ├── env.hcl
│ ├── logical/
│ │ └── terragrunt.hcl
│ └── physical/
│ └── terragrunt.hcl
└── production/
├── env.hcl
├── logical/
│ └── terragrunt.hcl
└── physical/
└── terragrunt.hcl
Similarly to region, this level has an env.hcl file that specifies input variables for the
individual environments. To see a full list of the variables available refer to the env.hcl file inside
any of the environment folders. All available variables are listed with descriptions and defaults.
The defaults included here are 3 separate environments:
- development
- A quickly deployed environment with all data protection and high availability disabled.
- Takes ~30 minutes to boot
- qa
- Same as development but with a different environment name.
- Takes ~30 minutes to boot
- production
- All data protections and high availability options enabled.
- Note: This stack will require manual intervention to remove deletion protections before deleting. We specifically set it up so our infrastructure code could not accidentally delete core resources.
- Takes ~45 minutes to boot
- All data protections and high availability options enabled.
Inside the module the only configuration you should ever need to modify is the infrastructure release version, you can see an example of the configuration block here:
terraform {
source = "git::https://github.com/Dozuki/CloudPrem-Infra.git//cloudprem/physical?ref=v2.6"
}At the end of the source= string you'll see a ?ref= parameter. This points to a specific release of our infrastructure code. If you want to upgrade your environment
to the latest infrastructure you'll need to modify this parameter to point to the desired release tag. See our main infrastructure release page to get
a list of the available tags. By default, this repo will always be updated to point to the latest available version but once you fork the repo you may need to make
these changes manually going forward.