AWS ParallelCluster 3.16.0 がリリースされ、診断ツール pcluster-diag が追加されたので試してみた

{
  "context": {
    "timestamp": "2026-08-22T03-52-04",
    "pcluster_diag_version": "1.0.0",
    "pcluster_version": "3.16.0",
    "instance_id": "i-088be9e32dcc38be0",
    "instance_type": "t3a.small",
    "node_type": "HeadNode",
    "cluster_config": {
      "AdditionalPackages": null,
      "AdditionalResources": null,
      "CustomS3Bucket": null,
      "DeploymentSettings": null,
      "DevSettings": null,
      "DirectoryService": null,
      "HeadNode": {
        "CustomActions": null,
        "Dcv": null,
        "DisableSimultaneousMultithreading": false,
        "Iam": {
          "AdditionalIamPolicies": [
            {
              "Policy": "arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore"
            }
          ],
          "InstanceProfile": null,
          "InstanceRole": null,
          "S3Access": null
        },
        "Image": null,
        "Imds": {
          "Secured": true
        },
        "InstanceType": "t3a.small",
        "LocalStorage": {
          "EphemeralVolume": null,
          "RootVolume": {
            "DeleteOnTermination": true,
            "Encrypted": true,
            "Iops": 3000,
            "Size": 45,
            "Throughput": 125,
            "VolumeType": "gp3"
          }
        },
        "Networking": {
          "AdditionalSecurityGroups": null,
          "ElasticIp": false,
          "Proxy": null,
          "SecurityGroups": null,
          "SubnetId": "subnet-029f0fb0acc64043d"
        },
        "SharedStorageEfsSettings": null,
        "SharedStorageType": "Ebs",
        "Ssh": {
          "AllowedIps": "0.0.0.0/0",
          "KeyName": null
        }
      },
      "Iam": null,
      "Image": {
        "CustomAmi": null,
        "Os": "ubuntu2404"
      },
      "Imds": {
        "ImdsSupport": "v2.0"
      },
      "LoginNodes": null,
      "Monitoring": {
        "Alarms": {
          "Enabled": true
        },
        "Dashboards": {
          "CloudWatch": {
            "Enabled": false
          }
        },
        "DetailedMonitoring": false,
        "Logs": {
          "CloudWatch": {
            "DeletionPolicy": "Delete",
            "Enabled": true,
            "RetentionInDays": 180
          },
          "Rotation": {
            "Enabled": true
          }
        }
      },
      "Region": "ap-northeast-1",
      "Scheduling": {
        "ScalingStrategy": "all-or-nothing",
        "Scheduler": "slurm",
        "SlurmQueues": [
          {
            "AllocationStrategy": "lowest-price",
            "CapacityReservationTarget": null,
            "CapacityType": "ONDEMAND",
            "ComputeResources": [
              {
                "CapacityReservationTarget": null,
                "CustomSlurmSettings": {},
                "DisableSimultaneousMultithreading": false,
                "DynamicNodePriority": 1000,
                "Efa": {
                  "Enabled": false,
                  "GdrSupport": false
                },
                "HealthChecks": {
                  "Gpu": {
                    "Enabled": null
                  }
                },
                "Instances": [
                  {
                    "InstanceType": "t3a.small"
                  }
                ],
                "LaunchTemplateOverrides": null,
                "MaxCount": 2,
                "MinCount": 1,
                "Name": "compute1",
                "Networking": {
                  "PlacementGroup": {
                    "Enabled": null,
                    "Id": null,
                    "Name": null
                  }
                },
                "SchedulableMemory": null,
                "SpotPrice": null,
                "StaticNodePriority": 1,
                "Tags": null
              }
            ],
            "ComputeSettings": {
              "LocalStorage": {
                "EphemeralVolume": null,
                "RootVolume": {
                  "Encrypted": true,
                  "Iops": 3000,
                  "Size": 45,
                  "Throughput": 125,
                  "VolumeType": "gp3"
                }
              }
            },
            "CustomActions": null,
            "CustomSlurmSettings": {},
            "HealthChecks": {
              "Gpu": {
                "Enabled": null
              }
            },
            "Iam": {
              "AdditionalIamPolicies": [
                {
                  "Policy": "arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore"
                }
              ],
              "InstanceProfile": null,
              "InstanceRole": null,
              "S3Access": null
            },
            "Image": null,
            "JobExclusiveAllocation": false,
            "Name": "queue1",
            "Networking": {
              "AdditionalSecurityGroups": null,
              "AssignPublicIp": null,
              "PlacementGroup": {
                "Enabled": null,
                "Id": null,
                "Name": null
              },
              "Proxy": null,
              "SecurityGroups": null,
              "SubnetIds": [
                "subnet-0897827eb4d8cd2fe"
              ]
            },
            "Tags": null
          }
        ],
        "SlurmSettings": {
          "CustomSlurmSettings": null,
          "CustomSlurmSettingsIncludeFile": null,
          "Database": null,
          "Dns": {
            "DisableManagedDns": false,
            "HostedZoneId": null,
            "UseEc2Hostnames": false
          },
          "EnableMemoryBasedScheduling": false,
          "ExternalSlurmdbd": null,
          "MungeKeySecretArn": null,
          "QueueUpdateStrategy": "COMPUTE_FLEET_STOP",
          "ScaledownIdletime": 5
        }
      },
      "SharedStorage": null,
      "Tags": [
        {
          "Key": "Name",
          "Value": "cluster-v3.16.0-pcluster-diag"
        },
        {
          "Key": "parallelcluster:cluster-name",
          "Value": "pcluster-diag-verify"
        },
        {
          "Key": "parallelcluster:version",
          "Value": "3.16.0"
        }
      ]
    },
    "dna_json": {
      "cluster": {
        "stack_name": "pcluster-diag-verify",
        "stack_arn": "arn:aws:cloudformation:ap-northeast-1:123456789012:stack/pcluster-diag-verify/0ca26650-9cfb-11f1-b77a-0663628fe01d",
        "raid_vol_ids": "",
        "raid_shared_dir": "",
        "raid_type": "",
        "base_os": "ubuntu2404",
        "region": "ap-northeast-1",
        "shared_storage_type": "ebs",
        "default_user_home": "shared",
        "efs_fs_ids": "",
        "efs_shared_dirs": "",
        "efs_encryption_in_transits": "",
        "efs_iam_authorizations": "",
        "efs_access_point_ids": "",
        "fsx_fs_ids": "",
        "fsx_mount_names": "",
        "fsx_dns_names": "",
        "fsx_volume_junction_paths": "",
        "fsx_fs_types": "",
        "fsx_shared_dirs": "",
        "volume": "",
        "scheduler": "slurm",
        "ephemeral_dir": "/scratch",
        "ebs_shared_dirs": "",
        "proxy": "NONE",
        "node_type": "HeadNode",
        "cluster_user": "ubuntu",
        "ddb_table": "parallelcluster-pcluster-diag-verify",
        "log_group_name": "/aws/parallelcluster/pcluster-diag-verify-202608210055",
        "dcv_enabled": "false",
        "dcv_port": "NONE",
        "enable_intel_hpc_platform": "false",
        "cw_logging_enabled": "true",
        "log_rotation_enabled": "true",
        "cluster_s3_bucket": "parallelcluster-50bb4ae12fec178e-v1-do-not-delete",
        "cluster_config_s3_key": "parallelcluster/3.16.0/clusters/pcluster-diag-verify-2jt60003s7xtui8o/configs/cluster-config-with-implied-values.yaml",
        "cluster_config_version": "umHZcAYU9ccWH1sCEk7qPuB_EOnAeNd_",
        "instance_types_data_version": "fBJZrI7ePvQtycamnKGJ38VyAkEK7fzm",
        "change_set_s3_key": "parallelcluster/3.16.0/clusters/pcluster-diag-verify-2jt60003s7xtui8o/configs/change-set.json",
        "instance_types_data_s3_key": "parallelcluster/3.16.0/clusters/pcluster-diag-verify-2jt60003s7xtui8o/configs/instance-types-data.json",
        "custom_node_package": "",
        "head_node_imds_secured": "true",
        "compute_node_bootstrap_timeout": 2100,
        "disable_sudo_access_for_default_user": "false",
        "launch_template_id": "HeadNodeLaunchTemplate",
        "dns_domain": "pcluster-diag-verify.pcluster.",
        "hosted_zone": "Z0247083T3YMGX1J5GQQ",
        "slurm_ddb_table": "parallelcluster-slurm-pcluster-diag-verify",
        "use_private_hostname": "false"
      }
    },
    "head_node_instance_id": "i-088be9e32dcc38be0"
  },
  "results": [
    {
      "check_id": "Imds",
      "check_description": "Verify that IMDS is responsive and functional.",
      "status": "WARNING",
      "warnings": [
        {
          "code": "W1",
          "message": "Instance tags metadata is not reachable via IMDS."
        }
      ]
    },
    {
      "check_id": "CfnHup",
      "check_description": "Verify that the cfn-hup daemon runs only on the head node and it is properly configured.",
      "status": "PASSED"
    },
    {
      "check_id": "ReservedUsersAndGroups",
      "check_description": "Verify that reserved ParallelCluster users and groups exist and do not share their uid/gid.",
      "status": "PASSED"
    },
    {
      "check_id": "CriticalPathsHaveExpectedPermissions",
      "check_description": "Verify that ParallelCluster critical files have their expected owner, group, and permissions.",
      "status": "PASSED"
    },
    {
      "check_id": "ClusterDaemonsAreRunning",
      "check_description": "Verify that the ParallelCluster management daemons for this node type are running.",
      "status": "PASSED"
    },
    {
      "check_id": "ClustermgtdHeartbeatIsHealthy",
      "check_description": "Verify that the clustermgtd heartbeat is fresh (clustermgtd is not stalled).",
      "status": "PASSED"
    },
    {
      "check_id": "DirectoryService",
      "check_description": "Verify that the directory service (NSS/SSSD/AD) integration is healthy.",
      "status": "SKIPPED_NOT_APPLICABLE"
    },
    {
      "check_id": "SlurmAccounting",
      "check_description": "Verify that Slurm accounting (slurmdbd, database, credentials, and configuration) is healthy.",
      "status": "SKIPPED_NOT_APPLICABLE"
    },
    {
      "check_id": "LustreFilesystem",
      "check_description": "Verify that FsxLustre filesystems are installed, mounted, reachable, and configured for EFA.",
      "status": "SKIPPED_NOT_APPLICABLE"
    }
  ]
}