Tested terraform version: v1.0.3
Before you run this:
-
(
az login
to the correct subscription before you begin) -
Edit 90-vars.tf to pick options
-
Create terraform.tfvars with username and password (example below)
username = "mydemouser"
password = "D0ntUseThisPassword"
- Run
terraform
(init
,plan
,apply
...) - Output will show you how to run the ansible playbook created
- Configure the VIP... original instructions
-
Some real docs on how to use this, current state of this README is more tracking what needs to be done
-
More variables , less hardcoded IP's.
-
Port2 Interface config differs from orignal script (Does it matter?)
-
UDR on the Jumpsubnet needs adjustment(?)
-
Configure and restrict outbound traffic with full inspection.
-
See comments in post.tpl
-
AKS creation taking forever can be a provsioning quota issue and/or the fortios template breaking and ending up with no config making AKS nodes unable to reach any network. Also note that
terraform destroy
after this kind of issue tends to get suck and deleting the resource group(s) manually might work faster. -
Permission for the AKS cluster LB can take some time, if you are quick with deploying the voteapp you might see this
Warning SyncLoadBalancerFailed 3m3s (x3 over 3m18s) service-controller Error syncing load balancer: failed to ensure load balancer: Retriable: false, RetryAfter: 0s, HTTPStatusCode: 403, RawError: Retriable: false, RetryAfter: 0s, HTTPStatusCode: 403, RawError: {"error":{"code":"AuthorizationFailed","message":"The client '..........
. -
Cloud-init can get stuck in apt(?!)
-
Adding the dynamic address in Real servers during provision seems to cause an issue, just removing it and adding it back again makes it work.