I have an Arch linux running windows 10 disk-based VMs. The disk is on a different volume group and is luks encrypted. I have a logical volume for each VM with ext4 file system. I manually edited the fstab with the correct UUIDs and I set the type to ext4. Before I installed windows on the VMs I rebooted to make sure the fstab was configured properly. After the installations, I'm getting this error for each partition right after I correctly type in the password for the disk:

[TIME] Timed out waiting for device /dev/disk/by-uuid/1bdc0382-d2a4-4581-b737-feec147dec40.
[DEPEND] Dependency failed for /disk0.
[DEPEND] Dependency failed for Local File Systems.
[DEPEND] Dependency failed for File System Check on /dev/disk/by-uuid/1bdc0382-d2a4-4581-b737-feec147dec40.

After those errors I get:

You are in emergency mode. After logging in type [...]

I'm not a linux expert so the answer might be simpler than it seems. Anyone has any suggestion?

EDIT #1:

fstab snippet:

# /dev/mapper/volgroup0-lv_disk0
UUID=1bdc0382-d2a4-4581-b737-feec147dec40   /disk0  ext4    rw,relatime 0 2

EDIT #2:

lsblk -f snippet:

NAME                FSTYPE      FSVER            LABEL       UUID                                   FSAVAIL FSUSE% MOUNTPOINTS
`-sda1              crypto_LUKS 2                            48bd9c70-c5cd-42c0-a58e-f0257be18d44
  `-disk            LVM2_member LVM2 001                     IVCIiW-5r2w-AzHY-hWyE-iJ7g-IqPB-lUdP9o

blkid snippet:

/dev/sda1: UUID="48bd9c70-c5cd-42c0-a58e-f0257be18d44" TYPE="crypto_LUKS" PARTUUID="fe7085b2-c19b-1f48-908c-c59dd96bcfc9"
/dev/mapper/disk: UUID="IVCIiW-5r2w-AzHY-hWyE-iJ7g-IqPB-lUdP9o" TYPE="LVM2_member"
/dev/mapper/volgroup0-lv_disk0: PTUUID="3421c065-23d3-48a1-8274-951444ce8d5c" PTTYPE="gpt"

EDIT #3:

fdisk -l snippet:

Disk /dev/mapper/disk: 447.12 GiB, 480086138368 bytes, 937668239 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
The primary GPT table is corrupt, but the backup appears OK, so that will be used.

Disk /dev/mapper/volgroup0-lv_disk0: 200 GiB, 214748364800 bytes, 419430400 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 3421C065-23D3-48A1-8274-951444CE8D5C

Device                                   Start       End   Sectors   Size Type
/dev/mapper/volgroup0-lv_disk0-part1      2048    206847    204800   100M EFI System
/dev/mapper/volgroup0-lv_disk0-part2    206848    239615     32768    16M Microsoft reserved
/dev/mapper/volgroup0-lv_disk0-part3    239616 418403031 418163416 199.4G Microsoft basic data
/dev/mapper/volgroup0-lv_disk0-part4 418404352 419426303   1021952   499M Windows recovery environment
The primary GPT table is corrupt, but the backup appears OK, so that will be used.
  • 1
    Can you clarify your disk setup? I understand that you have several logical volumes, each volume is LUKS encrypted, and there is a filesystem on each LUKS device. Is that correct? To understand better what is happening, disable the offending lines in /etc/fstab by commenting them and boot again. Add /etc/fstab and the output of lsblk -f as well as blkid to your question. By the way, both logical volumes and LUKS devices have persistent names out of the box, which you could use in /etc/fstab instead of the UUIDs. Names are easier to manage for humans than UUIDs. Jul 27 at 0:53
  • Ok will do that. To clarify: I have a separated HDD for the VMs. The disk has 1 partition which is luks encrypted. That partition has logical volumes for each VM. I have added one of the lvm mapping on the fstab to the question. Will do the rest in a minute. Jul 27 at 1:02
  • @berndbausch I have added snippets of the relevant parts because adding everything would pollute the post. If you need anything else, or for me to just dump everything let me know. Jul 27 at 1:26
  • 1
    I see that you LUKS-encrypted sda1, then used the LUKS device as a physical volume and created a logical volume on it. So far so good and correct. However, there doesn't seem to be a filesystem on that logical volume /dev/mapper/volgroup0-lv_disk0, which means that it can't be mounted. Where did you get the UUID 1bdc0382-d2a4-4581-b737-feec147dec40 from? Jul 27 at 1:50
  • @berndbausch I don't remember... I thought it was running fdisk -l but now I can't find it. I have edited the answer with an error shown in fdisk -l. EDIT: the other lvm on that disk does not show that same error. Jul 27 at 2:06

If you use a logical volume as a backing storage for virtual machines, the LV will be used "directly" as disk for the VM -- the ext4 filesystem you created was overwritten by the Windows installation so you can no longer mount it, because instead of ext4 your /dev/mapper/volgroup0-lv_disk0 LV now contains a partition table with Windows partitions.

If you want to access data from your Windows VM you can use libguestfs.

To fix your boot problem, remove the /dev/mapper/volgroup0-lv_disk0 entry from your fstab.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.