No nvme disk on boot

egoriv · 2021-11-22 16:23:56

(probably) After update, linux cant mount root partition because booter thing does not see my nvme ssd anymore. It sees connected usb and nvme is fully working and accessible from live environment. How can i fix it? Thank you.

That but longer and advancier:
Yesterday after rebooting arch after update on boot linux threw

mount: /new_root: can't find PARTUUID=(part uuid of an existing place)

i've dealt with this problem before (because i use efistub) so i run

mount /dev/nvme0n1p3 /new_root

(nvme0n1p3 is my root partition on my nvme drive) and was really surprised to it throwing an error, then i found out that there's no /dev/disk and no /dev/nvme0n1 and any of its partitions. Live environment detects that nvme, all of its partitions and files are there and i can successfully chroot in there, i've reinstalled and updated linux package (with /boot correctly mounted) but still.
After reboot initramfs (as i believe this mounting thing is part of) detected flash drive as /dev/sda but theres no nvme anywhere still.
---
I run generic amd laptop (tuf fx505dy), basically new,
basic kingston 512gb nvme
wiki efistub with amd ucode

Scimmia · 2021-11-22 16:26:28

This is probably the known bug in mkinitcpio. Manually add the vmd module to the initramfs (add it to the modules array in mkinitcpio.conf).

egoriv · 2021-11-22 16:54:06

Scimmia wrote:

This is probably the known bug in mkinitcpio. Manually add the vmd module to the initramfs (add it to the modules array in mkinitcpio.conf).

So adding vmd to /etc/mkinitcpio.conf on system root
so it looks like

/etc/mkinitcpio.conf
...
MODULES=(vmd)
...

And then

# mkinitcpio -P
...
--> Image generation was successful

then rebooting did nothing.
Did i do it right?

aFamiliarStranger · 2021-11-22 17:09:12

Are you in the current system? What I mean by that is if you are trying to mount the nvme from the ISO, and have not run the 'arch-chroot /mnt' command, then you are pointing the partition at a mount point which the system cannot locate when you are in the build. So in the case you are trying to mount from the ISO env. do:

mount /dev/nvme0n1p3 /mnt

Then

arch-chroot /mnt

Note: If it's the latter, then, you should have in total 3 partitions
/dev/nvme0n1p1 -- the disk partition
/dev/nvme0n1p2 -- the root partition
/dev/nmve0n1p3 -- the home partition

Finally, do a sanity check by running:

lsblk -l /dev/nvme0n1

and verify that that the output is something like what's shown below (Note: Depending on how you partitioned, it may look different than mine but the gist is about the same)

NAME      MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
nvme0n1   259:0    0 465.8G  0 disk
nvme0n1p1 259:1    0   512M  0 part /boot/efi
nvme0n1p2 259:2    0     4G  0 part [SWAP]
nvme0n1p3 259:3    0    50G  0 part /
nvme0n1p4 259:4    0 411.3G  0 part /home

If in the case you are trying to mount the partition as home directory, then, mount the root partition as shown above, but don't chroot yet. Do:

mkdir /mnt/new_root

mount /dev/nvme0n1p3 /mnt/new_root

MODULES=(amdgpu nvme)

If you are sure that your mount points are not faulty, then as shown above, adjust your modules and do

mkinitcpio -P

then update your boot-loader.

/etc/mkinitcpio.conf
# vim:set ft=sh
# MODULES
# The following modules are loaded before any boot hooks are
# run.  Advanced users may wish to specify all system modules
# in this array.  For instance:
#     MODULES=(piix ide_disk reiserfs)
MODULES=(amdgpu nvme)

# BINARIES
# This setting includes any additional binaries a given user may
# wish into the CPIO image.  This is run last, so it may be used to
# override the actual binaries included by a given hook
# BINARIES are dependency parsed, so you may safely ignore libraries
BINARIES=()

# FILES
# This setting is similar to BINARIES above, however, files are added
# as-is and are not parsed in any way.  This is useful for config files.
FILES=()

# HOOKS
# This is the most important setting in this file.  The HOOKS control the
# modules and scripts added to the image, and what happens at boot time.
# Order is important, and it is recommended that you do not change the
# order in which HOOKS are added.  Run 'mkinitcpio -H <hook name>' for
# help on a given hook.
# 'base' is _required_ unless you know precisely what you are doing.
# 'udev' is _required_ in order to automatically load modules
# 'filesystems' is _required_ unless you specify your fs modules in MODULES
# Examples:
##   This setup specifies all modules in the MODULES setting above.
##   No raid, lvm2, or encrypted root is needed.
#    HOOKS=(base)
#
##   This setup will autodetect all modules for your system and should
##   work as a sane default
#    HOOKS=(base udev autodetect block filesystems)
#
##   This setup will generate a 'full' image which supports most systems.
##   No autodetection is done.
#    HOOKS=(base udev block filesystems)
#
##   This setup assembles a pata mdadm array with an encrypted root FS.
##   Note: See 'mkinitcpio -H mdadm' for more information on raid devices.
#    HOOKS=(base udev block mdadm encrypt filesystems)
#
##   This setup loads an lvm2 volume group on a usb device.
#    HOOKS=(base udev block lvm2 filesystems)
#
##   NOTE: If you have /usr on a separate partition, you MUST include the
#    usr, fsck and shutdown hooks.
HOOKS=(base udev autodetect modconf numlock block filesystems keyboard fsck)

# COMPRESSION
# Use this to compress the initramfs image. By default, zstd compression
# is used. Use 'cat' to create an uncompressed image.
#COMPRESSION="zstd"

If none of the above works, then, I bet you did not add "-U" option when you generated fstab for the env. during installation. i.e

genfstab -U /mnt >> /mnt/etc/fstab

-Cheers!

V1del · 2021-11-22 17:17:04

If it used to work it's unlikely that VMD suddenly came up without an accompanying UEFI update nor is trying to add nvme to your modules, afaik it's a built-in.

From which kernel to which kernel did you update? Afaik there "were" some controllers broken by earlier 5.15 revisions and 5.15.4 contains a lot of reverts for broken patches, make sure you're at least on 5.15.4

Scimmia · 2021-11-22 17:19:48

I missed that it's an AMD laptop, so vmd doesn't make sense anyway. Sorry.

egoriv · 2021-11-22 17:45:57

V1del wrote:

If it used to work it's unlikely that VMD suddenly came up without an accompanying UEFI update nor is trying to add nvme to your modules, afaik it's a built-in.
From which kernel to which kernel did you update? Afaik there "were" some controllers broken by earlier 5.15 revisions and 5.15.4 contains a lot of reverts for broken patches, make sure you're at least on 5.15.4

pacman -Syu

got an linux update to 5.15.4 just like now )))

aFamiliarStranger wrote:

...
MODULES=(amdgpu nvme)
...
mkinitcpio -P
...

also did this
And one or both of this things helped, thank you all really much!!

aFamiliarStranger · 2021-11-22 17:56:11

Glad to have been helpful!

-Cheers!

ronanyeah · 2021-11-23 00:25:25

I also got burned by this today on an AMD laptop. I rolled back the linux and systemd packages I updated yesterday and added the modules above and ran mkinitcpio but no luck.

EDIT:

I came across this and rolling back to 5.14.16 fixed it for me: https://www.reddit.com/r/archlinux/comm … er_update/

Last edited by ronanyeah (2021-11-23 02:04:21)

Arch Linux

#1 2021-11-22 16:23:56

No nvme disk on boot

#2 2021-11-22 16:26:28

Re: No nvme disk on boot

#3 2021-11-22 16:54:06

Re: No nvme disk on boot

#4 2021-11-22 17:09:12

Re: No nvme disk on boot

#5 2021-11-22 17:17:04

Re: No nvme disk on boot

#6 2021-11-22 17:19:48

Re: No nvme disk on boot

#7 2021-11-22 17:45:57

Re: No nvme disk on boot

#8 2021-11-22 17:56:11

Re: No nvme disk on boot

#9 2021-11-23 00:25:25

Re: No nvme disk on boot

Board footer