CNI Terraform

cni-tf
├── Applications     --> CNI Kubernetes Applications
│   ├── Makefile
│   └── cni-inbound  --> CNI Inbound HelmChart Template Definitions
|   └── cni-outbound --> CNI Outbound HelmChart Template Definitions
├── Infrastructure
│   ├── deployer     --> Command Line Tool for deployment
│   └── provider     --> TF Infrastructure-as-code
├── Manifests
│   ├── Dev          --> CNI Deployment Manifests for Developers
│   └── Prod         --> CNI Deployment Manifests for Production
└── README.md

Requirements

Install Terraform:
```
brew install terraform
```
Install the AWS CLI:
```
brew install awscli
```
If it is already installed, ensure it is updated to the latest version:
```
brew upgrade awscli
```
Install kubectl:
```
brew install kubectl
```

If you have Docker installed, follow these additional steps (this will ensure your kubectl symlinks point to the brew installation):

rm /usr/local/bin/kubectl
brew link --overwrite kubernetes-cli

Install aws-iam-authenticator:
```
brew install aws-iam-authenticator
```
Install Helm:
```
brew install kubernetes-helm
```
Confirm your python alias points to python3
Install python packages:
```
pip install -r requirements.txt
```

Deploy a test environment

Make a copy of the base manifest file.

cp base_manifest.yaml test_manifest.yaml

Populate the empty values in cni_test.tfvars.json to reflect your desired setup. The table below details the meaning of each field.

Name	Description	Type
region	Name of the AWS region to deploy to	`string`
env	Name of the environment to deploy to	`string`
deployment_id	Short ID used to further distinguish a deployment. Must be less than 10 characters.	`string`
tags	Map of tags used to annotate each resource supporting tags	`map(string)`
lambda_layer_s3_key	Name of the lambda layer object in the S3 bucket	`string`
lambda_function_s3_key	Name of the lambda function object in the S3 bucket	`string`

Run the following command to perform a sanity check on your manifest:

python deployer/deploy.py -c plan -module provider/aws/cni_test/ -manifest -manifest test_manifest.yaml

If the previous step was successful, run this command to deploy the environment:

python deployer/deploy.py -c apply -module provider/aws/cni_test/ -manifest -manifest test_manifest.yaml

You should now have a running test environment.

deploy.py

Init

init configures the Terraform state backend for a given module using metadata from the provided manifest. It is needed to be run before running low-level commands such as terraform import or terraform state.

Validate

validate is a read-only command that verifies a provided -module (along with its nested modules) is syntactically valid. This check is purely static and does not take into account provided variables or existing state.

python deployer/deploy.py -c validate -module provider/aws/cni_test/ -manifest test_manifest.yaml

Plan

plan is a read-only command that outputs the exact list of changes Terraform will perform. This list of changes is stored as a terraform_plan file in the provided -module directory. This file is used by a successive apply command to perform the listed changes.

python deployer/deploy.py -c plan -module provider/aws/cni_test/ -manifest test_manifest.yaml

Apply

The apply command reads the terraform_plan file output by the previous plan run and performs each change that was listed. plan should always be run before apply.

python deployer/deploy.py -c apply -module provider/aws/cni_test/ -manifest test_manifest.yaml

Destroy

The destroy command destroys all infrastructure created by Terraform. This command will prompt for confirmation before destroying.

python deployer/deploy.py -c destroy -module provider/aws/cni_test/ -manifest test_manifest.yaml

Refresh

The refresh command updates Terraform's state to reflect real-world infrastructure. This is useful to detect drift from last-known state, such as in the event of an apply or destroy interruption.

python deployer/deploy.py -c destroy -module provider/aws/cni_test/ -manifest test_manifest.yaml

Deploy to EKS clusters

First, make sure your manifest is under Manifests/Dev. In the parent directory of this repository, run the following command:

make template

This will run helm to generate a number of k8s service files based on the contents of your manifest.

To submit services to a running EKS cluster, you will first need to update your local kubeconfig to point to it. Do this by running the following:

aws eks --region <REGION> update-kubeconfig --name <NAME OF EKS CLUSTER>

Run the next command to deploy the generated service files to the EKS cluster:

kubectl apply -f Manifest/Output/<RESOURCE_PREFIX>-<INBOUND OR OUTBOUND>-data-plane/cni-inbound/templates/

The EKS cluster should now be fully provisioned.

Note: in order too deploy to both the inbound and outbound EKS clusters, you'll need to repeat steps 2 and 3 for both inbound and outbound.

insta-code1/cni-deployer