IBM Cloud Satellite location on Azure

Use this terrafrom automation to set up satellite location on IBM cloud with azure host. It will provision satellite location and create 6 azure host and assign 3 host to control plane, and provision ROKS satellite cluster and auto assign 3 host to cluster, Configure cluster worker pool to an existing ROKS satellite cluster.

This is a collection of modules that make it easier to provision a satellite on IBM Cloud.

• satellite-location
• satellite-assign-host
• satellite-cluster
• satellite-cluster-worker-pool

Overview

IBM Cloud® Satellite helps you deploy and run applications consistently across all on-premises, edge computing and public cloud environments from any cloud vendor. It standardizes a core set of Kubernetes, data, AI and security services to be centrally managed as a service by IBM Cloud, with full visibility across all environments through a single pane of glass. The result is greater developer productivity and development velocity.

https://cloud.ibm.com/docs/satellite?topic=satellite-getting-started

Features

• Create satellite location.
• Create 6 azure host with RHEL 7-LVM.
• Assign the 3 hosts to the location control plane.
• Conditional creation:
  • Create a Red Hat OpenShift on IBM Cloud cluster and assign the 3 hosts to the cluster, so that you can run OpenShift workloads in your location.
  • Configure a worker pool to an existing OpenShift Cluster.

Compatibility

This module is meant for use with Terraform 0.13 or later.

Requirements

Terraform plugins

• Terraform 0.13 or later.
• terraform-provider-ibm

Install

Terraform

Be sure you have the correct Terraform version ( 0.13 or later), you can choose the binary here:

• https://releases.hashicorp.com/terraform/

Terraform provider plugins

Be sure you have the compiled plugins on $HOME/.terraform.d/plugins/

• terraform-provider-ibm

Usage

terraform init

terraform plan

terraform apply

terraform destroy

Example Usage

module "satellite-azure" {
  source = "git::[email protected]:terraform-ibm-modules/terraform-ibm-satellite-azure.git"

  ################# IBM cloud & Azure cloud variables #######################
  ibm_resource_group         = var.ibm_resource_group
  az_resource_group          = var.az_resource_group
  is_az_resource_group_exist = var.is_az_resource_group_exist
  az_region                  = var.az_region
  az_resource_prefix         = var.az_resource_prefix

  ################# Satellite location resource variables #######################
  is_location_exist    = var.is_location_exist
  location             = var.location
  managed_from         = var.managed_from
  location_zones       = var.location_zones
  location_bucket      = var.location_bucket
  host_labels          = var.host_labels
  satellite_host_count = var.satellite_host_count
  addl_host_count      = var.addl_host_count

  ################# Satellite Cluster resource variables #######################
  create_cluster             = var.create_cluster
  cluster                    = var.cluster
  cluster_host_labels        = var.cluster_host_labels
  create_cluster_worker_pool = var.create_cluster_worker_pool
  worker_pool_name           = var.worker_pool_name
  worker_pool_host_labels    = var.worker_pool_host_labels
  create_timeout             = var.create_timeout
  update_timeout             = var.update_timeout
  delete_timeout             = var.delete_timeout
}

Note

• satellite-location module creates new location or use existing location ID/name to process. If user pass the location which is already exist, satellite-location module will error out and exit the module. In such cases user has to set is_location_exist value to true. So that module will use existing location for processing.
• satellite-location module download attach host script to the $HOME directory and appends respective permissions to the script.
• satellite-location module will update the attach host script and will be used in the custom_data attribute of azurerm_linux_virtual_machine resource.

Inputs

Name	Description	Type	Default	Required
ibm_resource_group	Resource Group Name that has to be targeted.	string	n/a	yes
is_az_resource_group_exist	"If false, resource group (az_resource_group) will be created. If true, existing resource group (az_resource_group) will be read"	bool	false	no
az_resource_group	Azure Resource Group	string	satellite-azure	no
az_region	Azure Region	string	eastus	yes
location	Name of the Location that has to be created	string	n/a	satellite-azure
is_location_exist	Determines if the location has to be created or not	bool	false	no
managed_from	The IBM Cloud region to manage your Satellite location from.	string	wdc	no
location_zones	Allocate your hosts across three zones for higher availablity	list	[]	no
host_labels	Add labels to attach host script	list	[env:prod]	no
location_bucket	COS bucket name	string	n/a	no
host_count	The total number of host to create for control plane. host_count value should always be in multiples of 3, such as 3, 6, 9, or 12 hosts	number	3	yes
addl_host_count	The total number of additional host	number	3	no
az_source_address_prefix	Azure security group source address prefix	list(string)	[*]	no
az_security_custom_rules	Azure security group custom rules	list(object)	n/a	no
azure_vnet_address_space	Azure vNet address space	list(string)	["10.0.0.0/16"]	no
azure_vnet_subnet_prefixes	Azure vNet subnet prefixes	list(string)	["10.0.1.0/24", "10.0.2.0/24", "10.0.3.0/24"]	no
az_cp_vm_os_disk	Azure control plane vm os disk	object	n/a	no
az_cp_vm_source_image_reference	Azure control plane vm source image reference	object	n/a	no
az_cluster_vm_os_disk	Azure cluster vm os disk	object	n/a	no
az_cluster_vm_source_image_reference	Azure cluser vm source image reference	object	n/a	no
resource_prefix	Prefix to the Names of all VSI Resources	string	satellite-azure	no
public_key	Public SSH key used to provision Host/VSI	string	n/a	no
location_instance_type	Profile information of location hosts	string	Standard_D4s_v3	no
cluster_instance_type	Profile information of cluster hosts	string	Standard_D4s_v3	no
cluster	Name of the ROKS Cluster that has to be created	string	n/a	yes
cluster_zones	Allocate your hosts across these three zones	set	n/a	yes
kube_version	Kuber version	string	n/a	yes
default_wp_labels	Labels on the default worker pool	map	n/a	no
workerpool_labels	Labels on the worker pool	map	n/a	no
cluster_tags	List of tags for the cluster resource	list	n/a	no
create_cluster_worker_pool	Create Cluster worker pool	bool	false	no
worker_pool_name	Worker pool name	string	satellite-worker-pool	no
workerpool_labels	Labels on the worker pool	map	n/a	no
create_timeout	Timeout duration for creation	string	n/a	no
update_timeout	Timeout duration for updation	string	n/a	no
delete_timeout	Timeout duration for deletion	string	n/a	no

Outputs

Name	Description
location_id	Location id
host_script	Host registartion script content
network_security_group_id	Network security group Id
network_security_group_name	Network security group name
floating_ip_addresses	Floating IP Addresses
vnet_id	The Id of the newly created vNet
vnet_name	The Name of the newly created vNet
control_plane_private_ips	The Primary Private IP Address assigned to control plane VMs
control_plane_public_ips	The Primary Public IP Address assigned to control plane VMs
cluster_private_ips	The Primary Private IP Address assigned to cluster VMs
cluster_public_ips	The Primary Public IP Address assigned to cluster VMs
cluster_worker_pool_id	Cluster worker pool id
worker_pool_worker_count	worker count deatails
worker_pool_zones	workerpool zones

Pre-commit Hooks

Run the following command to execute the pre-commit hooks defined in .pre-commit-config.yaml file

pre-commit run -a

We can install pre-coomit tool using

pip install pre-commit

How to input varaible values through a file

To review the plan for the configuration defined (no resources actually provisioned)

terraform plan -var-file=./input.tfvars

To execute and start building the configuration defined in the plan (provisions resources)

terraform apply -var-file=./input.tfvars

To destroy the VPC and all related resources

terraform destroy -var-file=./input.tfvars

All optional parameters by default will be set to null in respective example's varaible.tf file. If user wants to configure any optional paramter he has overwrite the default value.