Pod manager counts all virsh pools as available disk
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
MAAS |
Won't Fix
|
High
|
Unassigned |
Bug Description
On my laptop virsh is configured to have access to two pools which are two different directories on my system. I added virsh as a pod to MAAS but it thinks my system has double the amount of disk space. MAAS should recognize that both pools are on the same disk and only count it once.
$ df
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/mapper/
/dev/sda2 241965 138970 90503 61% /boot
/dev/sda1 121982 3496 118486 3% /boot/efi
$ maas maas pods read
Success.
Machine-readable output follows:
[
{
"id": 1,
},
"name": "suited-gopher",
],
"used": {
},
"type": "virsh",
],
"total": {
}
}
]
$ virsh pool-list
Name State Autostart
-------
default active yes
ISOs active yes
$ virsh pool-info default
Name: default
UUID: f2bb24f2-
State: running
Persistent: yes
Autostart: yes
Capacity: 441.95 GiB
Allocation: 267.05 GiB
Available: 174.90 GiB
$ virsh pool-info ISOs
Name: ISOs
UUID: 109ba190-
State: running
Persistent: yes
Autostart: yes
Capacity: 441.95 GiB
Allocation: 267.05 GiB
Available: 174.90 GiB
Changed in maas: | |
assignee: | nobody → Newell Jensen (newell-jensen) |
importance: | Undecided → High |
Changed in maas: | |
status: | Incomplete → Triaged |
Changed in maas: | |
milestone: | 2.2.1 → none |
How are we supposed to recognize this? Looking at the virsh information you provided they both look like different pools? We can only perform virsh commands so it must be noticed in virsh not outside of virsh in anyway.
I think the real question is why do you have 2 pools pointing to the same backing? This seems more like a bad configuration in virsh then a real bug in MAAS.