Merge pull request #12 from ryanmcafee/feature/terraform-workspace-support

Feature/terraform workspace support

Author: ryanmcafee

.gitignore

@@ -7,4 +7,5 @@ credentials
 terraform.tfvars
 terraform.tfstate.d
 backend.hcl
-customizations
+customizations
+terraform.tfstate

README.md

@@ -6,9 +6,9 @@ Flavorize aims to make it easy to setup and provision a fully working Kubernetes
 
 To setup/provision a Kubernetes cluster, follow the documentation for the appropriate provider. The following guides make it possible to setup a Kubernetes cluster in under 1 hr, on your pick of cloud provider. Your mileage will vary based on your knowledge and experience, but I have aimed to make the setup as streamlined as possible to help encourage the adoption and usage of Kubernetes and other cloud native technologies. Flavorize is now baked into my workflow and can't imagine using Kubernetes without it. I have built kubernetes "flavors" for Azure as well as Digital Ocean of which you may find quick starts on how to use below.
 
-[Azure](docs/providers/Azure.md)        
+[Azure](flavors/azure/readme.md)        
 
-[Digital Ocean](docs/providers/DigitalOcean.md)
+[Digital Ocean](flavors/digitalocean/readme.md)
 
 # Contributing
 

destroy.ps1

@@ -0,0 +1,19 @@
+#! /usr/bin/pwsh
+
+[CmdletBinding()]
+Param(
+    [string]$flavor = "azure",
+    [string]$tf_workspace = (Invoke-Expression -Command "terraform workspace show").Trim(),
+    [string]$kube_config_dir = "credentials/workspaces/${tf_workspace}/.kube",
+    [Parameter(Position=0,Mandatory=$false,ValueFromRemainingArguments=$true)]
+    [string[]]$Args
+)
+
+#Initialize terraform
+terraform init flavors/${flavor}
+
+$provider_specific_variable_file=If (Test-Path "customizations/workspaces/${tf_workspace}/provider.tfvars" -PathType Leaf) {"customizations/workspaces/${tf_workspace}/provider.tfvars"} Else {"customizations/provider.tfvars"}
+$flavor_variable_file=If (Test-Path "customizations/workspaces/${tf_workspace}/flavorize.tfvars" -PathType Leaf) {"customizations/workspaces/${tf_workspace}/flavorize.tfvars"} Else {"customizations/flavorize.tfvars"}
+
+#Create the terraform plan
+terraform destroy -var-file="${provider_specific_variable_file}" -var-file="${flavor_variable_file}" flavors/${flavor}

docs/Contributing.md

@@ -4,17 +4,17 @@
 
 In order to add support for an additional cloud provider you'll want to do the following:
 
-1) Add your new Terraform cloud provider resource plugin to providers.tf and make sure that you set the correct version number for it to work reliably for others.
+1) Create an issue for the provider you are adding support for
 
-2) Create additional directory under modules/k8s/{cloudProviderName} and place 3 files there: k8s.tf, variables.tf and output.tf
+2) Clone the repo locally
 
-3) Consult documentation for your cloud provider that you are adding support for and update k8s.tf in your directory accordingly. Please also ensure that the provider's resources will only attempt provisioning when the value for your cloud_provider is set to your new provider. You can accomplish this by using a ternary statement and the count property. Look at the azure provider for an example of this.
+3) Create a feature branch from master
 
-4) Ensure that the available output variables conform to the same output variables used in: modules/k8s/azure/output.tf so we can continue to interchangably provision common resources across cloud providers (certmanager, ingress controller, externaldns, etc..).
+4) Navigate to the "flavors" directory and expand it.
 
-5) Add any additional provider specific variables to: modules/k8s/{cloudProviderName}/variables.tf.
+5) Copy the "flavors/azure" folder into "flavors/yournewflavor"
 
-6) Add your new provider to main.tf in the directory root with the same module name of "k8s". After doing so and testing it out to make sure it works, please comment out the additional provider. To my knowledge, Terraform currently doesn't offer a way to disable a module at the module level, so for now this is the best work around I can think of. Please open an issue, if you have a better recommendation.
+6) Update the following files in the new flavor folder: k8s.tf (provider specific kubernetes cluster provisioning and calls the flavorize module), output.tf (store the kubernetes cluster info in tf state), providers.tf (terraform provider setup), variables.tf (variables specific to your new provider)
 
 7) Lastly, ensure that any additional variables specific to your provider have been added to terraform.tfvars.example