Windows 11 VM won't boot, no disks, drvload works, but after restart disks gone again

[Adamg]

Hi all,

I have a Windows 11 VM, it crashed on the weekend and was going to automatic windows recovery on every reboot. I restored a backup from Friday, but it still goes to the windows recovery options (none of which fix anything of course).

I entered cmd and used diskpart and realized that there are no disks visible to windows, but they are there, I ran the command:

drvload d:\vioscsi\2k19\amd64\vioscsi.inf

And that made the 2TB C drive show up, then I tried to click the windows option "Exit and Continue to Windows"

But that restarts the VM, and goes back to recovery options and again NO DISKS.

I found a proxmox forum where they said to ALSO run:

dism /image:C:\ /add-driver /driver:\vioscsi\2k19\amd64

to install the driver to the C drive, which I tried also, but still after restarting the VM it goes back to the recovery screen and there are no disks until I manually run drvload.

There must be something wrong with this VM because the other forums said those 2 commands fixed it.

Any ideas?

 

[Maximiliano]

Hello,

Could you please share your VM config with us? You can find it at `/etc/pve/qemu-server/<VM_ID>.conf`.

 

[Adamg]

For sure! Here it is:

bios: ovmf
boot: order=scsi0;ide0;net0
cores: 8
cpu: x86-64-v2-AES
efidisk0: local-zfs:vm-105-disk-3,efitype=4m,pre-enrolled-keys=1,size=1M
ide0: local:iso/virtio-win-0.1.240.iso,media=cdrom,size=612812K
machine: pc-q35-8.1
memory: 12288
meta: creation-qemu=8.1.2,ctime=1701907028
name: WONT-BOOT-Mar-4-2024-Win11-Pro
net0: virtio=BC:2B:11:BB:79:FB,bridge=vmbr0,firewall=1
numa: 0
ostype: win11
scsi0: local-zfs:vm-105-disk-0,discard=on,iothread=1,size=2000G
scsihw: virtio-scsi-single
smbios1: uuid=ae69bce3-0cf9-4479-83e4-d1cbc162dd63
sockets: 1
tpmstate0: local-zfs:vm-105-disk-2,size=4M,version=v2.0
vmgenid: 3ad27c11-5efb-495a-b65b-355069ba3551c

------------

I already solved it by reloading a dump backup from 2 weeks ago as a different VM #, it booted fine, then I moved the disk from the unbootable VM to the working VM, used linux and rsync to update the files, BUT I'd still like to know how to fix the problem because it might happen again.

Thanks!

 

[sybord]

Hello,

I have the same issue so if anyone know how to fix this i'm interested !

Thanks

 

[emunt6]

Old problem, simple fix:

For Linux:

discard=off
aio=native
cache=none
scsihw=virtio-scsi-pci

For Windows:

discard=off
aio=native
cache=none

For Windows dont use any "virto" emulation, you will hit by many bugs.

 

[Maximiliano]

Do note that we recommend to use VirtIO both on Linux and Windows in general. In Windows' case, you need to properly install the drivers before using them [1] and, indeed, failing to do so would result in a disk that is not recognized by the OS.

Regarding Discard, it is also recommended in general, but it can be problematic with some very old hardware or software so it is worth testing with this option disabled.

[1] https://pve.proxmox.com/wiki/Windows_VirtIO_Drivers