DiskMB property of the HTTP API Job Spec is not having any effect on disk requirement

[ncodes]

Nomad version

Nomad v0.5.4

Operating system and Environment details

Ubuntu 16.04.1 LTS (GNU/Linux 4.4.0-64-generic x86_64)

Issue

I am currently creating jobs through the HTTP API and I noticed that the disk resource is always set to 0 in the Task Resource section of nomad alloc-status .... Running nomad inspect ... on the job shows EphemeralDisk.SizeMB of task group still set to the default 300 even when DiskMB of the job spec is set to a different value. I looked at the source and found TaskGroups.EphemeralDisk , applied it, ran nomad alloc-status ... and got the expected value. But the output of nomad alloc-status ... still showed Disk to be 0.

Reproduction

curl -X POST http://localhost:4646/v1/jobs -d '{ ... }'

Nomad Client logs (if appropriate)

ID                  = 0ed83e2d
Eval ID             = be1e1274
Name                = 1834baa0-d9c1-4a4a-b63c-57ae59941778.1834baa0-d9c1-4a4a-b63c-57ae59941778[0]
Node ID             = 4cd92524
Job ID              = 1834baa0-d9c1-4a4a-b63c-57ae59941778
Client Status       = running
Client Description  = <none>
Desired Status      = run
Desired Description = <none>
Created At          = 03/05/17 17:24:48 UTC

Task "task-1834baa0-d9c1-4a4a-b63c-57ae59941778" is "running"
Task Resources
CPU        Memory          Disk  IOPS  Addresses
4/100 MHz  58 MiB/512 MiB  0 B   0     connector: 10.132.0.2:33769

Recent Events:
Time                   Type                   Description
03/05/17 17:24:54 UTC  Started                Task started by client
03/05/17 17:24:48 UTC  Driver                 Downloading image ...
03/05/17 17:24:48 UTC  Downloading Artifacts  Client is downloading artifacts
03/05/17 17:24:48 UTC  Received               Task received by client`

Job file (if appropriate)

This is the initial job spec without the EphemeralDisk task group property.

  "Job": {
    "Region": "global",
    "ID": "task",
    "Name": "task",
    "Type": "service",
    "Priority": 50,
    "AllAtOnce": false,
    "Datacenters": [
      "dc1"
    ],
    "Constraints": [
      {
        "LTarget": "${attr.kernel.name}",
        "RTarget": "linux",
        "Operand": "="
      }
    ],
    "TaskGroups": [
      {
        "Name": "task",
        "Count": 1,
        "Constraints": null,
        "Tasks": [
          {
            "Name": "task-1",
            "Driver": "docker",
            "Config": {
              "image": "golang:1.7.5-wheezy",
              "force_pull": true,
              "command": "bash",
              "network_mode": "host",
              "args": [""],
              "privileged": true,
              "port_map": []
            },
            "Env": {
            },
            "Services": [],
            "Meta": {
            },
            "LogConfig": {
              "MaxFiles": 10,
              "MaxFileSizeMB": 10
            },
            "Templates": [],
            "Artifacts": [],
            "Resources": {
              "CPU": 500,
              "MemoryMB": 512,
              "IOPS": 0,
              "Networks": [
                {
                  "MBits": 200,
                  "DynamicPorts":[
                    { "Label":"connector"}
                  ]
                }
              ]
            },
            "DispatchPayload": {}
          }
        ],
        "Resources": {
          "CPU": 500,
          "MemoryMB": 512,
          "DiskMB": 430,
          "IOPS": 0,
          "Networks": []
        },
        "RestartPolicy": {
          "Interval": 300000000000,
          "Attempts": 10,
          "Delay": 25000000000,
          "Mode": "delay"
        },
        "Meta": {}
      }
    ],
    "Update": {
      "Stagger": 10000000000,
      "MaxParallel": 1
    }
  }
}

This one includes the EphemeralDisk property

  "Job": {
    "Region": "global",
    "ID": "task",
    "Name": "task",
    "Type": "service",
    "Priority": 50,
    "AllAtOnce": false,
    "Datacenters": [
      "dc1"
    ],
    "Constraints": [
      {
        "LTarget": "${attr.kernel.name}",
        "RTarget": "linux",
        "Operand": "="
      }
    ],
    "TaskGroups": [
      {
        "Name": "task",
        "Count": 1,
        "Constraints": null,
        "Tasks": [
          {
            "Name": "task-1",
            "Driver": "docker",
            "Config": {
              "image": "golang:1.7.5-wheezy",
              "force_pull": true,
              "command": "bash",
              "network_mode": "host",
              "args": [""],
              "privileged": true,
              "port_map": []
            },
            "Env": {
            },
            "Services": [],
            "Meta": {
            },
            "LogConfig": {
              "MaxFiles": 10,
              "MaxFileSizeMB": 10
            },
            "Templates": [],
            "Artifacts": [],
            "Resources": {
              "CPU": 500,
              "MemoryMB": 512,
              "IOPS": 0,
              "Networks": [
                {
                  "MBits": 200,
                  "DynamicPorts":[
                    { "Label":"connector"}
                  ]
                }
              ]
            },
            "DispatchPayload": {}
          }
        ],
        "Resources": {
          "CPU": 500,
          "MemoryMB": 512,
          "DiskMB": 430,
          "IOPS": 0,
          "Networks": []
        },
       "EphemeralDisk": {
          "Sticky": false,
          "Migrate": false,
          "SizeMB": 200
        },
        "RestartPolicy": {
          "Interval": 300000000000,
          "Attempts": 10,
          "Delay": 25000000000,
          "Mode": "delay"
        },
        "Meta": {}
      }
    ],
    "Update": {
      "Stagger": 10000000000,
      "MaxParallel": 1
    }
  }
}

[dadgar]

@ncodes Thanks for catching that! Will be fixed in 0.5.5

[github-actions]

I'm going to lock this issue because it has been closed for 120 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.