Total pod resources not matching used pod resources

[Newell Jensen]

Total pod resources are not matching used pod resources:

# maas admin pods read
Success.
Machine-readable output follows:
[
    {
        "total": {
            "memory": 249984,
            "local_storage": 959999999999,
            "cores": 352,
            "local_disks": 7
        },
        "used": {
            "memory": 262144,
            "local_storage": 960272990208,
            "cores": 352,
            "local_disks": 8
        },
        "resource_uri": "/MAAS/api/2.0/pods/213/",
        "available": {
            "memory": -12160,
            "local_storage": -272990209,
            "cores": 0,
            "local_disks": -1
        },
        "id": 213,
        "architectures": [
            "amd64/generic"
        ],
        "name": "hot-gecko",
        "capabilities": [
            "composable",
            "fixed_local_storage"
        ],
        "type": "rsd"
    }
]

[Newell Jensen (newell-jensen)]

With some investigation it was found that this bug is a result of the commissioning process that MAAS performs (related to bug #1659917). To test this, I pre-composed all the machines (eight) in the RSD Pod. I then created the pod via the MAAS API, which at the end prints out the pod information. Since commissioning hasn't been completed at this point, we can see that there is not an issue with the information that is presented. However, I then commissioned all the nodes and on completion, I read the pod information from the MAAS API again. This time, the information is out of sync and doesn't match.

root@maas-dev:~# maas admin pods create type="rsd" power_address="10.3.0.1:8443" power_user="admin" power_pass="admin"
Success.
Machine-readable output follows:
{
    "total": {
        "memory": 249984,
        "local_storage": 959999999999,
        "cores": 352,
        "local_disks": 8
    },
    "used": {
        "memory": 249984,
        "local_storage": 959999999992,
        "cores": 352,
        "local_disks": 8
    },
    "resource_uri": "/MAAS/api/2.0/pods/219/",
    "available": {
        "memory": 0,
        "local_storage": 7,
        "cores": 0,
        "local_disks": 0
    },
    "id": 219,
    "architectures": [
        "amd64/generic"
    ],
    "name": "model-lemur",
    "capabilities": [
        "composable",
        "fixed_local_storage"
    ],
    "type": "rsd"
}
root@maas-dev:~# maas admin pods read
Success.
Machine-readable output follows:
[
    {
        "available": {
            "local_disks": 0,
            "local_storage": -272990209,
            "cores": 0,
            "memory": -12160
        },
        "name": "model-lemur",
        "resource_uri": "/MAAS/api/2.0/pods/219/",
        "id": 219,
        "total": {
            "local_disks": 8,
            "cores": 352,
            "local_storage": 959999999999,
            "memory": 249984
        },
        "type": "rsd",
        "used": {
            "local_disks": 8,
            "cores": 352,
            "local_storage": 960272990208,
            "memory": 262144
        },
        "capabilities": [
            "composable",
            "fixed_local_storage"
        ],
        "architectures": [
            "amd64/generic"
        ]
    }
]

[Newell Jensen (newell-jensen)]

This is still a bug for the local_disks being out of sync but the memory and local_storage issue still needs to be corrected.