Qcow2 image size read improperly

[modsbyus]

Hi folks,

CPU(s) 24 x Intel(R) Xeon(R) CPU E5645 @ 2.40GHz (2 Sockets)

Kernel Version Linux 5.15.53-1-pve #1 SMP PVE 5.15.53-1 (Fri, 26 Aug 2022 16:53:52 +0200)

PVE Manager Version pve-manager/7.2-7/d0dd0e85

I have a VM with a 75 GB qcow2 image that is 75 GB but Proxmox is reading this file as 68.72 TB. If I check in CLI, I see
ls -l
total 72130840
-rw-r----- 1 root root 75838455808 Oct 31 09:17 vm-115-disk-0.qcow2
I have tried shutting down this VM, and re-importing but it did the same thing. I really need to fix this as it is a production Windows server and I can't run backups on it or migrate the disk.
Let me know any other info you need.
Thanks

 

[bbgeek17]

What does the output look like for: qemu-img info [filename] ?
What does the VM show internally for its disk size?

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[modsbyus]

image: vm-115-disk-0.qcow2
file format: qcow2
virtual size: 62.5 TiB (68719476736000 bytes)
disk size: 68.8 GiB
cluster_size: 65536
Format specific information:
compat: 1.1
compression type: zlib
lazy refcounts: false
refcount bits: 16
corrupt: false
extended l2: false

internal disk shows 64 TB. See picture. Two unallocated partitions. One is 1992.66 GB and the other is 61952 GB plus the system 350Mb partition and the OS partition 55GB.

 

[bbgeek17]

virtual size: 62.5 TiB (68719476736000 bytes)
This ^ shows where the sizing is coming from. You could have a corrupted qcow. It could be a sparse empty space. It could be snapshots.

What does the VM show internally for its disk size?

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[modsbyus]

From Windows disk management, disk shows 64 TB. See picture in previous reply. Two unallocated partitions. One is 1992.66 GB and the other is 61952 GB plus the system 350Mb partition and the OS partition 55GB.
I think its just empty space but I don't know how to get rid of it.

 

[bbgeek17]

search keywords that should help: "qcow2 reduce size"
This was the first result for me: https://pve.proxmox.com/wiki/Shrink_Qcow2_Disk_Files

If this is a production machine, I would not be doing any experiments on it. Copy or clone it somewhere and test your procedure thoroughly.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[gfngfn256]

Let me know any other info you need.

I know this is really an old post, and it looks like the OP has gone AWOL. But I've noticed that in the OP's original post image, he's showing what appears as the same virual disk: vm-115-disk-0.qcow2 as both scsi0 & scsi1. They appear to have different nfs servers. Maybe some elaborate mirroring going on, IDW. But possibly linked to OP's problem size.

 

[bbgeek17]

vm-115-disk-0.qcow2 as both scsi0 & scsi1. They appear to have different nfs servers. Maybe some elaborate mirroring going on, IDW. But possibly linked to OP's problem size.

Disk's index counter across different storage pools starts from 0, so what you see is completely normal.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox