EKS version 1.19 unable to find ami

[emilchitas]

I have issues

I'm submitting a...

• bug report
• feature request
• support request - read the FAQ first!
• kudos, thank you, warm fuzzy

What is the current behavior?

When trying to deploy an EKS cluster, version 1.19, I get the following error:
Error: Your query returned no results. Please change your search criteria and try again.
It seems as if the module is unable to find a passing ami for the nodes.

If this is a bug, how to reproduce? Please include a code sample if relevant.

Deploy a version 1.19 EKS cluster

module "eks" {
  source                          = "terraform-aws-modules/eks/aws"
  version                         = "14.0.0"
  cluster_name                    = "${var.name}-${var.environment}-cluster"
  cluster_version                 = "1.19"
  subnets                         = var.private-subnets
  vpc_id                          = var.vpc_id
  cluster_endpoint_private_access = true
  cluster_endpoint_public_access  = false
  enable_irsa                     = true
  write_kubeconfig                = false

  node_groups = {
    default = {
      name             = "${var.name}-${var.environment}-node-group"
      min_capacity     = 1
      desired_capacity = var.eks_node_group.desired_capacity
      max_capacity     = var.eks_node_group.max_capacity
      instance_types   = [var.eks_node_group.instance_type]
      key_name         = var.keypair_name
      tags = [
        {
          "key"                 = "k8s.io/cluster-autoscaler/enabled"
          "propagate_at_launch" = "false"
          "value"               = "true"
        },
        {
          "key"                 = "k8s.io/cluster-autoscaler/${local.cluster_name}"
          "propagate_at_launch" = "false"
          "value"               = "true"
        }
      ]
    }
  }
  map_roles = local.roles
  map_users = local.users
  tags      = local.common_tags
}

What's the expected behavior?

Terraform should be able to find a suitable ami for this cluster version.

Are you able to fix this problem and submit a PR? Link here if you have already.

No

Environment details

• Affected module version: 14.0.0
• OS: Ubuntu
• Terraform version: 0.14.6

Any other relevant info

None

[alghanmi]

I am able to reproduce this issue.

As far as I can tell, there are no EKS optimized windows AMIs for Kubernetes 1.19. The latest available is for Kubernetes 1.18.

Since I do not use Windows workers in EKS, I added the following variable in module block to resolve the issue:

worker_ami_name_filter_windows = "Windows_Server-2019-English-Core-EKS_Optimized-1.18-*"

[jmcgeheeiv]

When you implement this, you can also remove the code in local.tf that tests for Kubernetes >= 1.14, as the minimum EKS Kubernetes version is now 1.15:

  # Windows nodes are available from k8s 1.14. If cluster version is less than 1.14, fix ami filter to some constant to not fail on 'terraform plan'.
  worker_ami_name_filter_windows = (var.worker_ami_name_filter_windows != "" ?
    var.worker_ami_name_filter_windows : "Windows_Server-2019-English-Core-EKS_Optimized-${tonumber(var.cluster_version) >= 1.14 ? var.cluster_version : 1.14}-*"
  )

[barryib]

I just opened #1371 to address this. Will you please review it and test it ?

[newb1e]

I just opened #1371 to address this. Will you please review it and test it ?

To upgrade from 1.19 to 1.20 I ran with the module from the PR and the issue with the windows AMI is resolved.
Upgrade and TF run finished successfully

[nahidupa]

Looks like in cluster_version: 1.20 it's broken again, with the following filter getting an error.

worker_ami_name_filter_windows= "Windows_Server-2004-English-Core-EKS_Optimized-1.20*"
worker_ami_owner_id_windows = "amazon"

Error: Your query returned no results. Please change your search criteria and try again.

@alghanmi alghanmi

[barryib]

This has been fixed in the latest version of this module.

Please upgrade your module and follow changelog and docs/upgrades.md for more info.

[nahidupa]

@barryib thanks for the reply I have tried with the latest master branch. Looks like it cannot locate the images with filter criteria, I have checked AWS the images is exist.
worker_ami_name_filter_windows= "Windows_Server-2004-English-Core-EKS_Optimized-1.20*"

Another question is can I create a worker group with directly assign AMI ?

The filter(worker_ami_name_filter_windows) is good, However, in production, without a test changing AMI can cause an issue.

[github-actions]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues. If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.