Description of problem:
Upstream bug: https://bugs.launchpad.net/nova/+bug/1469179

Nova seems to be reporting local disk for NFS backed cinder volumes. This is a problem trying to create VMs in the overcloud. Found found references to this bug for juno, NOT kilo, then it seems to be back for liberty.

https://ask.openstack.org/en/question/65854/incorrect-advertisement-of-block-storage-total-and-free-space/?answer=92450#post-id-92450
https://ask.openstack.org/en/question/88772/if-i-boot-from-a-volume-cinder-why-appears-as-im-using-computes-local-disk-usage/
https://bugs.launchpad.net/nova/+bug/1469179

~~~
[stack@controller1 ~]$ nova hypervisor-stats
+----------------------+---------+
| Property             | Value   |
+----------------------+---------+
| count                | 2       |
| current_workload     | 0       |
| disk_available_least | 396     |
| free_disk_gb         | -300    |
| free_ram_mb          | 950218  |
| local_gb             | 400     |
| local_gb_used        | 700     |
| memory_mb            | 1031114 |
| memory_mb_used       | 80896   |
| running_vms          | 14      |
| vcpus                | 64      |
| vcpus_used           | 25      |
+----------------------+---------+
~~~

~~~
[heat-admin@compute1 ~]$ df -h
Filesystem                    Size  Used Avail Use% Mounted on
/dev/sda2                     279G   26G  254G  10% /
devtmpfs                      252G     0  252G   0% /dev
tmpfs                         252G   54M  252G   1% /dev/shm
tmpfs                         252G  1.4M  252G   1% /run
tmpfs                         252G     0  252G   0% /sys/fs/cgroup
tmpfs                          51G     0   51G   0% /run/user/991
10.29.104.17:/osp7cf_glance   500G   23G  478G   5% /var/lib/glance/images
tmpfs                          51G     0   51G   0% /run/user/990
10.29.104.27:/osp7cf_cinder3  820G  1.5G  818G   1% /var/lib/cinder/mnt/7476a0c43f2f41a0b2b5b39afaf2f095
10.29.104.15:/osp7cf_cinder2  820G  8.4G  811G   2% /var/lib/cinder/mnt/fbdef210720e36d661073a6e8ee114ab
10.29.104.15:/osp7cf_cinder1  820G   21G  799G   3% /var/lib/cinder/mnt/5ea493792ed71f306f40d94699619c1a
10.29.104.27:/osp7cf_cinder4  820G  1.4G  818G   1% /var/lib/cinder/mnt/57128760568a05d1c0e3e78b6ed64817
tmpfs                          51G     0   51G   0% /run/user/1000
~~~

Additional info:
From the upstream bug report:
~~~
instance.root_gb means size of host local disk used by the instance root disk. All code that uses this attribute treats it that way.

Volume-backed instances have root disks placed in Cinder. Therefore obviously root_gb should be 0 for them. However currently this is not so. The same applies to min_disk and size image attributes, used for boot from a volume.

As a result the code which uses these attributes works incorrectly. Some problems are already detected [1], other are not yet [2].

There are two kinds of bugs:
1 Nova fails to launch an instance from a large volume if the volume size (or an original image's min_disk) is greater than requested flavor.root_gb.
2 Nova incorrectly calculates host disk space consumed by volume-backed instances.

To fix all these problems fully it is proposed to set root_gb, min_disk, size to 0 for volume-backed instances.

[1] https://bugs.launchpad.net/nova/+bug/1334974
https://bugs.launchpad.net/nova/+bug/1459491
https://bugs.launchpad.net/nova/+bug/1466305
https://bugs.launchpad.net/nova/+bug/1457517
https://bugs.launchpad.net/nova/+bug/1358566
[2] https://github.com/openstack/nova/blob/master/nova/notifications.py#L407
https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L250
~~~