Hey there,
I've got a few questions / issues regarding a nvme ssd passed through in a VM.
The server has 2 nvmes - one as LVM storage for VM disks, another (Seagate BarraCuda Q5 ZP10000CV30001) passed in a VM.
The VM is Windows 11, installed directly onto that nvme as ovmf Q35 ***without*** an efi disk.
Sometimes the VM enters into EFI shell after reboot (usually when I add new hardware / hdd passed and reboot it) and won't boot at all until the whole server is restarted.
Once the server restarts - the VM boots/reboots normally w/o any issues whatsoever.
I think the issues with the VM unable to boot are due to the NVME drive itself, but I can't be sure (because I also don't have an EFI disk for that VM).
I booted the VM using Gparted-Live ISO, and noted the following nvme errors:
nvme0n1: I/O Cmd(0x2) @ LBA 264, 248 blocks, I/O Error (sct 0x2 / sc 0x82) MORE
protection error, dev nvme0n1, sector 264 op 0x0
READ) flags 0x80700 phys_seg 16 prio class 2
(there were more of these for different blocks / sectors / etc, but I didn't take a picture)
Has anyone encountered such issues? Do you think the nvme itself is to blame (it's few months old), or something else? Possible solution?
I'm still running PVE6 (yes, I know it's EOL), kernel 5.11.22-5-pve
Also, If this is a known issue that is addressed on pve7, do let me know. I'd rather not have to upgrade just yet (other reasons).
Here's the VM config:
Regards,
D
I've got a few questions / issues regarding a nvme ssd passed through in a VM.
The server has 2 nvmes - one as LVM storage for VM disks, another (Seagate BarraCuda Q5 ZP10000CV30001) passed in a VM.
The VM is Windows 11, installed directly onto that nvme as ovmf Q35 ***without*** an efi disk.
Sometimes the VM enters into EFI shell after reboot (usually when I add new hardware / hdd passed and reboot it) and won't boot at all until the whole server is restarted.
Once the server restarts - the VM boots/reboots normally w/o any issues whatsoever.
I think the issues with the VM unable to boot are due to the NVME drive itself, but I can't be sure (because I also don't have an EFI disk for that VM).
I booted the VM using Gparted-Live ISO, and noted the following nvme errors:
nvme0n1: I/O Cmd(0x2) @ LBA 264, 248 blocks, I/O Error (sct 0x2 / sc 0x82) MORE
protection error, dev nvme0n1, sector 264 op 0x0
READ) flags 0x80700 phys_seg 16 prio class 2(there were more of these for different blocks / sectors / etc, but I didn't take a picture)
Has anyone encountered such issues? Do you think the nvme itself is to blame (it's few months old), or something else? Possible solution?
I'm still running PVE6 (yes, I know it's EOL), kernel 5.11.22-5-pve
Also, If this is a known issue that is addressed on pve7, do let me know. I'd rather not have to upgrade just yet (other reasons).
Here's the VM config:
Code:
balloon: 0
bios: ovmf
boot: order=sata2;hostpci0
cores: 16
cpu: host,flags=+md-clear;-ibpb;-virt-ssbd;-amd-ssbd;+amd-no-ssb;+pdpe1gb;+hv-tlbflush;+aes
cpuunits: 262143
hostpci0: 0000:48:00,rombar=0 # the nvme
hostpci1: 0000:01:00,pcie=1,x-vga=1 # rtx 3090
hostpci2: 0000:44:00.1,pcie=1,rombar=0 # usb controller
hostpci3: 44:00.3,pcie=1,rombar=0 # usb controller
hostpci4: 0000:21:00,pcie=1 # NIC
hugepages: 1024
machine: pc-q35-5.2
memory: 32768
name: Windows11
net0: virtio=86:49:17:1D:EA:28,bridge=vmbr0,tag=555
numa: 1
onboot: 1
ostype: win10
sata2: none,media=cdrom
scsi2: /dev/disk/by-id/ata-ST5000DM000-1FK178_W4J04EPK,backup=0,cache=writeback,iothread=1,size=4883770584K
scsi3: /dev/disk/by-id/ata-ST12000NM000J-2TY103_ZRT0EG3Z,backup=0,cache=writeback,iothread=1,size=11176G
scsihw: virtio-scsi-single
smbios1: uuid=eeeb9398-9d7e-48b0-b27c-8315717f96e2
sockets: 1
startup: order=2,up=60
vga: none
vmgenid: b22af7d5-adb4-4dee-a075-a861dec43616Regards,
D