Can you passthrough a known bad HD to a VM?

[CancunManny]

I have a known bad 4TB HD. I tried to rescue files with my regular PC but was just too cumbersome since I had to keep computer on, had case open with hard drive stiking out, etc.

I now have a new computer, so my old one became a proxmox server. I created a VM with ubuntu, and was able to passthrough a good 4TB hard drive that I plan to use to capture the recovered files. However when I try to passthrough the known bad hard drive I get an error. I want to use tools like ddrescue, testdisk, etc to try and recover some files.

When I list disks this is what I get

root@proxmox:~# find /dev/disk/by-id/ -type l|xargs -I{} ls -l {}|grep -v -E '[0-9]$' |sort -k11|cut -d' ' -f9,10,11,12 16:32 /dev/disk/by-id/ata-ST4000DM000_WDH0ZXXX -> ../../sda 16:32 /dev/disk/by-id/wwn-0x5000c5009d4b4XXX -> ../../sda 16:32 /dev/disk/by-id/ata-KINGSTON_SV300S37A240G_50026B7246055XXX -> ../../sdb 16:32 /dev/disk/by-id/wwn-0x50026b7246055XXX -> ../../sdb 19:01 /dev/disk/by-id/ata-ST4000DM004-2U9104_ZW60PXXX -> ../../sdc 19:01 /dev/disk/by-id/wwn-0x5000c500e6c82XXX -> ../../sdc root@proxmox:~#

sda is the known bad hard drive
sdb is the hd being used by proxmox
sdc is the 4tb drive that will receive the recovered files

As a side note, I only passed the /dev/disk/by-id/wwn-0x5000c500e6c82XXX -> ../../sdc to my VM, do I also need to passthrough /dev/disk/by-id/ata-ST4000DM004-2U9104_ZW60PXXX -> ../../sdc or should I be passing only the ata one?

When I try to passthrough the bad drive I get this:


root@proxmox:~# qm set 104 -scsi2 /dev/disk/by-id/ata-ST4000DM000_WDH0ZA0M update VM 104: -scsi2 /dev/disk/by-id/ata-ST4000DM000_WDH0ZA0M qemu-img: Could not open '/dev/disk/by-id/ata-ST4000DM000_WDH0ZA0M': Could not read image for determining its format: Input/output error could not parse qemu-img info command output for '/dev/disk/by-id/ata-ST4000DM000_WDH0ZA0M' - malformed JSON string, neither tag, array, object, number, string or atom, at character offset 0 (before "(end of string)") at /usr/share/perl5/PVE/Storage/Plugin.pm line 954. volume /dev/disk/by-id/ata-ST4000DM000_WDH0ZA0M does not exist root@proxmox:~#

How can I access the bad drive to scan it and try to recover files?

Thanks in advance for any and all suggestions

 

[Dunuin]

Thats no real passthrough. For your VM to be able to directly access the real physical disks you would need to PCI passthrough a whole disk controller (with all disks attached to it). With "qm set" your VM is working with a virtual disk that is mapped to the physical disk so there is another abstraction layer in between you probably don't want when trying to rescue data from a bad disk.

Would be easier to boot some data recovery Live linux distro bare metal and then try with that to rescue some data.

 

[CancunManny]

Thanks for the feedback. Seems I will store the bad drive until I have a bare metal I can sacrifice for a week or two while I try to recover some files.

ty again!