suyambuganesh82 / aks-creator-fleet-ship

AKS Creator

This tool image allows you to create and manage the configuration files of a AKS Kubernetes repository.

Getting started

In order to be able to use this image, it must be started in the desired AKS Cluster repository using Docker or Podman. Furthermore, the repository directory must be mounted in the container. To avoid permission errors, it is recommended to do this with the permissions of the executing user.

Igor

We will use the igor. The tool opens a shell in your favorite docker container mounting your current workspace into the container.

The following steps are executed:

• Start the image
• Mount necessary directories
• Set permissions
• Clean up

To install igor just download the igor.sh and store it in your $PATH like this:

sudo curl https://raw.githubusercontent.com/la-cc/igor/main/igor.sh -o /usr/local/bin/igor
sudo chmod +x /usr/local/bin/igor

Configure igor

Running igor without configuration will launch a busybox image. In order to use the tool with the AKS Creator image, a configuration file is required.

NOTE: You can get the recent tag from la-cc/aks-creator-core

Open the file .igor.sh in your preferred editor and use the following settings to configure igor:

# select docker image
IGOR_DOCKER_IMAGE=ghcr.io/la-cc/aks-creator-core:0.0.2
IGOR_DOCKER_COMMAND=                                  # run this command inside the docker container
IGOR_DOCKER_PULL=0                                    # force pulling the image before starting the container (0/1)
IGOR_DOCKER_RM=1                                      # remove container on exit (0/1)
IGOR_DOCKER_TTY=1                                     # open an interactive tty (0/1)
IGOR_DOCKER_USER=$(id -u)                             # run commands inside the container with this user
IGOR_DOCKER_GROUP=$(id -g)                            # run commands inside the container with this group
IGOR_DOCKER_ARGS=''                                   # default arguments to docker run
IGOR_PORTS=''                                         # space separated list of ports to expose
IGOR_MOUNT_PASSWD=1                                   # mount /etc/passwd inside the container (0/1)
IGOR_MOUNT_GROUP=1                                    # mount /etc/group inside the container (0/1)
IGOR_MOUNTS_RO="${HOME}/.azure"                       # space separated list of volumes to mount read only
IGOR_MOUNTS_RW=''                                     # space separated list of volumes to mount read write
IGOR_WORKDIR=${PWD}                                   # use this workdir inside the container
IGOR_WORKDIR_MODE=rw                                  # mount the workdir with this mode (ro/rw)
IGOR_ENV=''                                           # space separated list of environment variables set inside the container

Workflow

The following workflow is recommended as part of a aks cluster creation.

No.	Step	required	Tool
0	Allow you configure the azure backend to save the remote state for terraform.	no	config-azure-backend
1	Initialize empty configuration file.	yes	config-init
2	Fill missing fields in configuration file.	yes	-
3	Template the whole aks platform folder structure.	yes	config-template
4	Create the AKS Cluster with needed context.	yes	terrform (version >= 1.3.0")

0. Create a Azure Backend for Terraform State (Optional) config.yaml

Requirements:

• Azure Active Directory Access
• Azure Subscription Access

The easiest way is to fill the file .backend.env with the necessary values.

Then execute the script (from inside the aks-creator-core container):

config-azure-backend

You can also start the script with interactive mode:

config-azure-backend -i or config-azure-backend --interactive

1. Initialize empty configuration file

Requirements: none

Create an empty config.yaml. This already contains the necessary structure and placeholders for the values, as required in the following step. To do so simply execute the script (from inside the aks-creator-core container):

config-init

You will get a config.yaml that can be filled by you.

2. Fill missing fields in config.yaml

Requirements:

• config.yaml of 1. 1. Get configuration file config.yaml

You can get more information from 00. Configuration Options for config.yaml

3. Template the AKS folder structure from config.yaml

Requirements:

• config.yaml of 1. 1. Get configuration file config.yaml

To do so simply execute the script (from inside the aks-creator-core container):

config-template

4. Terraform Apply

Requirements:

• Azure Active Directory Access
• Azure Subscription Access

4.1 Terraform Apply + Azure Backend

If go through the step 0. Create a Azure Backend for Terraform State (Optional) then you need to execute the following commands (from inside the aks-creator-core container or local terraform binary):

terraform init
terraform select workspace <STAGE>
terraform plan -var-file=env/<STAGE>/terraform.tfvars

If the plan is fine for you, then apply it with:

terraform apply -var-file=env/<STAGE>/terraform.tfvars -auto-approve

4.2 Terraform Apply + Local Backend

If you don't create azure backend then execute the following commands (from inside the aks-creator-core container or local terraform binary):

terraform init
#terraform version < 1.4.*
terraform workspace new <STAGE>
#terraform version  >= 1.4.*
terraform workspace select -or-create <STAGE>
terraform plan -var-file=env/<STAGE>/terraform.tfvars

If the plan is fine for you, then apply it with:

terraform apply -var-file=env/<STAGE>/terraform.tfvars -auto-approve

00. Configuration Options for config.yaml

The following examples show the possible configuration of the templating. The used module itself can be further adjusted or overwritten.

Minimum necessary configuration

---
# Azure Backend for Terraform related data
azure_backend: {}

# Azure Kubernetes Cluster related data
clusters:
  - name: <valiant>
    stage: <development>
    azure_public_dns: {}
    node_pools: {}

# Azure Devops Pipeline related data
azure_devops_pipeline: {}

Maximum possible configuration:

---
# Azure Backend for Terraform related data
azure_backend:
  enable: true
  resource_group_name_backend: <"rg-tfstate-backend-example">
  storage_account_name: <"satfstatebackendexample">

# Azure Kubernetes Cluster related data
clusters:
  - name: <valiant>
    stage: <development>
    admin_list: ["8a7sdwd", "47243892"]
    azure_public_dns:
      enable: true
      azure_cloud_zone: <"your-domain.de">
    node_pool_count: <3>
    vm_size: <"Standard_B4ms">
    kubernetes_version: <"1.24.9">
    orchestrator_version: <"1.24.9">
    node_pools:
      enable_node_pools: true
      pool:
        - name: <"internal">
          min_count: <1>
          max_count: <3>
          node_count: <2>

# Azure Devops Pipeline related data
azure_devops_pipeline:
  enable: <true>
  library_group: <tkc-fleet>