terraform-provider-aws-parallelcluster fails on parallelcluster 3.11.0 with login nodes enabled

[kondakovm]

The terraform-provider-aws-parallelcluster fails while parsing the cluster status during creation on the 3.11 API with login nodes enabled, resulting in the following error:

 Error: Error while waiting for cluster to finish updating.

   with module.parallelcluster_clusters.aws-parallelcluster_cluster.managed_configs["ParallelCluster"],
   on .terraform/modules/parallelcluster_clusters/modules/clusters/main.tf line 35, in resource "aws-parallelcluster_cluster" "managed_configs":
   35: resource "aws-parallelcluster_cluster" "managed_configs" {

 json: cannot unmarshal array into Go struct field _DescribeClusterResponseContent.loginNodes of type map[string]interface {}

Despite this error, the cluster was created and is fully operational, but terraform cannot read or import it ending with the same error. This is most likely connected to transitioning from a single login node to multiple login nodes in a pool.

Additional info: The deployment with login nodes works on parallelcluster 3.10.1. The deployment works on parallelcluster 3.11.0 without login nodes enabled.

Required Info:

• AWS ParallelCluster version 3.11.0
• Full cluster configuration without any credentials or personal data:

  Region: eu-central-1
  CustomS3Bucket: parallelcluster-custom-bucket-name
  Image:
  Os: alinux2
  SharedStorage:
  - MountDir: /home
    Name: parallelcluster_shared
    StorageType: Ebs
    EbsSettings:
      VolumeType: gp3
      Size: 1000
      DeletionPolicy: Delete
  HeadNode:
  InstanceType: c5.large
  LocalStorage:
    RootVolume:
      Size: 100
      VolumeType: gp3
      DeleteOnTermination: true
  Networking:
    SubnetId: subnet-123456789123
  Iam:
    AdditionalIamPolicies:
      - Policy: arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore
  Ssh:
    KeyName: parallelcluster_ssh_key
  LoginNodes:
  Pools:
    - Name: login
      Count: 1
      InstanceType: t3.small
      Ssh:
        KeyName: parallelcluster_ssh_key
      Networking:
        SubnetIds:
          - subnet-123456789123
      Iam:
        AdditionalIamPolicies:
          - Policy: arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore
  Scheduling:
  Scheduler: slurm
  SlurmQueues:
    - Name: queue1
      CapacityType: SPOT
      Networking:
        SubnetIds:
          - subnet-123456789123
      Iam:
        AdditionalIamPolicies:
          - Policy: arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore
      ComputeResources:
        - InstanceType: c5.xlarge
          MinCount: 0
          MaxCount: 10
          Name: c5xlarge
  SlurmSettings:
    QueueUpdateStrategy: TERMINATE
    Dns:
      DisableManagedDns: true
      UseEc2Hostnames: true

[hanwen-pcluste]

Thank you for reporting the issue. We will work on a fix

[hanwen-pcluste]

The problem is solved in ParallelCluster 3.11.1. Please use the latest version.

[kondakovm]

Thank you for taking care of the issue, unfortunately, I got the same error when deploying the config with the 3.11.1 API using the parallelcluster provider:

Error: Error while waiting for cluster to finish updating.
  with module.parallelcluster_clusters.aws-parallelcluster_cluster.managed_configs["ParallelCluster"],
  on .terraform/modules/parallelcluster_clusters/modules/clusters/main.tf line 35, in resource "aws-parallelcluster_cluster" "managed_configs":
  35: resource "aws-parallelcluster_cluster" "managed_configs" {
json: cannot unmarshal array into Go struct field _DescribeClusterResponseContent.loginNodes of type map[string]interface {}

The cluster is created and fully functional, no errors in Lambda API logs, but terraform can't read/modify login nodes status.

[gmarciani]

Hi @kondakovm ,

we are working on the issue with 3.11.1. Will give an update there once we have more info.

Thank you for reporting the issue.

[gmarciani]

Currently working on the fix for the next provider version https://github.com/aws-tf/terraform-provider-aws-parallelcluster/pull/206