You are not logged in.
Hello,
I updated my system and after that I could not restart my laptop normally. I was greeted by this error message when booting.
Here you can see the output of the update after which the error occurred.
After that I had no time to take care of the problem for a week. Then I tried to update the system again, hoping that this would fix the problem by itself. I did this by booting into the arch iso, logging into my system with arch-chroot and then running
pacman -Syu
. In doing so, I got this output, which looks completely normal to me.
But before I did that I printed the output of journalctl.
After I rebooted my laptop regularly , it started in emergency mode (Here is the output).
Here is the output of
journalctl -xb
.
Information about my System:
- Hardware: Lenovo IdeaPad Flex 5 14ALC05 (Arch Wiki Page for my device)
- I have installed the packages
amd-ucode 20230404.2e92a49f-1
and
xf86-video-amdgpu 23.0.0-1
- I use LUKS disk encryption for my root and home partition
Unfortunately I have no idea what the problem is and would be very grateful for help!
I would be happy to provide more information that could help solve the problem.
Last edited by archforthewin (2023-06-22 08:37:12)
Offline
Jun 04 17:27:16 archlaptop mount[603]: mount: /boot: unbekannter Dateisystemtyp »vfat«.
…
Jun 04 17:27:15 archlaptop ufw-init[440]: modprobe: FATAL: Module ip_tables not found in directory /lib/modules/6.3.4-arch1-1
Compare
pacman -Q linux
and
uname -a
If the versions (likely) don't match you forgot to mount the /boot partition before the update.
Sidebar: please never use "-x" w/ journalctl, it's effectively spam.
Offline
Oh yes thank you so much for pointing that out! I have indeed forgotten to mount /boot.
Output of
uname -a
:
Linux archiso 5.19.6-arch-1 #1 SMP PREEMPT_DYNAMIC Wed, 31 Aug 2022 22:09:40 +0000 x85_64 GNU/Linux
Output of
pacman -Q linux
:
linux 6.3.5.arch1-1
The two versions do not match. But the first one also refers to the ISO I booted from and the second one to my actual system I arch-chrooted into. Why would they match?
How do I fix this problem? Is it safe to mount /boot this time and update again? Should that solve this problem?
Offline
But the first one also refers to the ISO I booted from and the second one to my actual system I arch-chrooted into. Why would they match?
Of course not, that would only work in the emergency shell.
Boot the install iso, arch-chroot into the system, mount the /boot partition and re-install the kernel.
Offline
Yes that makes sense.
I followed your instructions and reinstalled the kernel. That worked, so I didn't end up in the emergency shell at boot time.
The system update seems to have solved my original problem so my system now boots normally.
This is the current output of
journalctl -rb -p 3
These error messages, some of which also refer to amdgpu were apparently present before my problem and do not seem to have been responsible for the problem. I don't know what exactly the problem was, but the system update seems to have fixed it.
Very big thanks to you for the quick help. I find it remarkable and great how many people you have already helped with your knowledge.
Thank you very much!
Offline
Hello,
after my initial error went away after the system update, which was performed in a roundabout way, it has now reappeared after a reboot (without me having updated my system again afterwards).
I have no idea what exactly this error refers to and would be very grateful for help.
Last edited by archforthewin (2023-06-09 15:17:05)
Offline
There is no error in that output on that image, what makes you think there's an error? That's just general output the amdgpu driver generates during init, none of these look like an error.
Last edited by V1del (2023-06-09 15:22:21)
Online
My laptop displays this message and will not boot up further. If these are not error messages I wonder what the error is and why just this message is issued here.
Offline
I just noticed that a /boot directory exists even before I mount the boot partition. Is this normal? Can this be because I tried to fix the error by running an update without mounting the boot partition first and files were created in /boot? Should I change something about this?
Offline
It is normal to have a /boot directory, it should just be empty and I suspect it's not for you.
Re-install the kernel like the last time and post your /etc/fstab
Offline
You're right. For me the /boot directory is not empty. Is it advisable to delete the contents manually?
Here is my /etc/fstab.
Offline
I have arch-chrooted into my system and run
mount -a
. But when I run
pacman -S linux
I get the following error:
error: failed to init transaction (unable to lock database)
error: clould not lock database: Read-only file system
.
The last time I reinstalled the kernel I did not use mount -a but only mounted the boot (and home) partition manually. Other than that, I don't think I did anything different.
It seems that my root partition is mounted in read only mode (Output of "findmnt --target /var/lib/pacman" confirms that). But why is that the case?
I found out that it had something to do with my hard drive encryption. I have now successfully reinstalled the kernel. In fact, my laptop is now booting successfully again. However, it is not a solution to the problem if I have to reinstall the kernel just because I want to reboot my laptop. Does anyone have any idea why reinstalling the kernel would do anything and/or how I can really fix the problem?
Last edited by archforthewin (2023-06-09 16:57:44)
Offline
Ftr, your fstb
# Static information about the filesystems.
# See fstab(5) for details.
# <file system> <dir> <type> <options> <dump> <pass>
# /dev/mapper/vg1-root
UUID=85771b69-d4fc-4bb1-bc01-d09fc030c3d8 / ext4 rw,relatime 0 1
# /dev/mapper/vg1-home
UUID=dae6144e-5866-4ed2-ae45-7967970b3e60 /home ext4 rw,relatime 0 2
# /dev/nvme0n1p1
UUID=CE47-0146 /boot vfat rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,utf8,errors=remount-ro 0 2
# /dev/mapper/vg1-swap
UUID=02e192e2-060d-46ba-a5d3-b78cbc8cd633 none swap defaults 0 0
Please post the output of
lsblk -f
mount
file /boot/vmlinuz-linux
(and please here; not in a mile-long gist)
Offline
These are the requested outputs:
NAME FSTYPE FSVER LABEL UUID FSAVAIL FSUSE% MOUNTPOINTS
nvme0n1
├─nvme0n1p1 vfat FAT32 CE47-0146 115,3M 41% /boot
└─nvme0n1p2 crypto_LUKS 2 afa2d6f3-ce37-44de-99a0-0e66cfad1b76
└─cryptlvm LVM2_member LVM2 001 OLjPl1-fggO-dJ0z-xcO0-7wjp-sdV9-rS0FyS
├─vg1-root ext4 1.0 85771b69-d4fc-4bb1-bc01-d09fc030c3d8 3,8G 87% /
├─vg1-swap swap 1 02e192e2-060d-46ba-a5d3-b78cbc8cd633 [SWAP]
└─vg1-home ext4 1.0 dae6144e-5866-4ed2-ae45-7967970b3e60 155,2G 57% /home
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
sys on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
dev on /dev type devtmpfs (rw,nosuid,relatime,size=3726236k,nr_inodes=931559,mode=755,inode64)
run on /run type tmpfs (rw,nosuid,nodev,relatime,mode=755,inode64)
efivarfs on /sys/firmware/efi/efivars type efivarfs (rw,nosuid,nodev,noexec,relatime)
/dev/mapper/vg1-root on / type ext4 (rw,relatime)
securityfs on /sys/kernel/security type securityfs (rw,nosuid,nodev,noexec,relatime)
tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev,inode64)
devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000)
cgroup2 on /sys/fs/cgroup type cgroup2 (rw,nosuid,nodev,noexec,relatime,nsdelegate,memory_recursiveprot)
pstore on /sys/fs/pstore type pstore (rw,nosuid,nodev,noexec,relatime)
bpf on /sys/fs/bpf type bpf (rw,nosuid,nodev,noexec,relatime,mode=700)
systemd-1 on /proc/sys/fs/binfmt_misc type autofs (rw,relatime,fd=34,pgrp=1,timeout=0,minproto=5,maxproto=5,direct,pipe_ino=1823)
mqueue on /dev/mqueue type mqueue (rw,nosuid,nodev,noexec,relatime)
hugetlbfs on /dev/hugepages type hugetlbfs (rw,relatime,pagesize=2M)
debugfs on /sys/kernel/debug type debugfs (rw,nosuid,nodev,noexec,relatime)
tracefs on /sys/kernel/tracing type tracefs (rw,nosuid,nodev,noexec,relatime)
fusectl on /sys/fs/fuse/connections type fusectl (rw,nosuid,nodev,noexec,relatime)
configfs on /sys/kernel/config type configfs (rw,nosuid,nodev,noexec,relatime)
ramfs on /run/credentials/systemd-sysusers.service type ramfs (ro,nosuid,nodev,noexec,relatime,mode=700)
ramfs on /run/credentials/systemd-sysctl.service type ramfs (ro,nosuid,nodev,noexec,relatime,mode=700)
ramfs on /run/credentials/systemd-tmpfiles-setup-dev.service type ramfs (ro,nosuid,nodev,noexec,relatime,mode=700)
tmpfs on /tmp type tmpfs (rw,nosuid,nodev,size=3738412k,nr_inodes=1048576,inode64)
/dev/mapper/vg1-home on /home type ext4 (rw,relatime)
/dev/nvme0n1p1 on /boot type vfat (rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,utf8,errors=remount-ro)
ramfs on /run/credentials/systemd-tmpfiles-setup.service type ramfs (ro,nosuid,nodev,noexec,relatime,mode=700)
tmpfs on /run/user/1000 type tmpfs (rw,nosuid,nodev,relatime,size=747680k,nr_inodes=186920,mode=700,uid=1000,gid=1000,inode64)
gvfsd-fuse on /run/user/1000/gvfs type fuse.gvfsd-fuse (rw,nosuid,nodev,relatime,user_id=1000,group_id=1000)
portal on /run/user/1000/doc type fuse.portal (rw,nosuid,nodev,relatime,user_id=1000,group_id=1000)
/boot/vmlinuz-linux: Linux kernel x86 boot executable bzImage, version 6.3.5-arch1-1 (linux@archlinux) #1 SMP PREEMPT_DYNAMIC Tue, 30 May 2023 13:44:01 +0000, RO-rootFS, swap_dev 0XB, Normal VGA
These outputs were NOT created by chrooting into my system, but after a normal boot, after the kernel reinstall a boot was possible again.
For me as a recommendation for the future: should short outputs like this be embedded in the post itself and only long outputs like from large log files be provided as a gist?
Last edited by archforthewin (2023-06-10 11:38:40)
Offline
The config looks ok and the boot partition is automounted by the fstab and the kernel image there is the current one.
Before you run the next kernel update, check
mount | grep boot
to make sure the boot partition is (still) mounted.
You can also
sudo umount /boot
and check whether there're files in /boot (not supposed to be the case)
Fyi, your root partition is pretty full, you might want to clear the pacman cache?
Offline
Thanks for the info regarding my boot partition space. I am aware of this and regularly clean the pacman cache and the docker build cache which are responsible for this.
Since my boot partition is listed in fstab, it is normally mounted naturally. This was just not the case when I chrooted into my system to run the update when I could not boot normally. There I forgot to execute "mount -a".
Accordingly, as mentioned in post #11, there are files in /boot when the boot partition is not mounted.
❯ sudo umount /boot
❯ ls /boot/
initramfs-linux-fallback.img initramfs-linux.img vmlinuz-linux
~
❯ sudo mount /boot
~
❯ ls /boot/
amd-ucode.img EFI grub initramfs-linux-fallback.img initramfs-linux.img vmlinuz-linux
What is the best way to deal with this? Should I delete these files manually?
Last edited by archforthewin (2023-06-10 13:09:08)
Offline
Should I delete these files manually?
Yes, they've no business being there.
You can also try to protect mountpoint by making the directory read-only and/or immutable to cause errors when anything attempts to write there
sudo umount /boot
sudo chattr +i /boot
sudo touch /boot/foo # failure
sudo mount /boot
sudo touch /boot/bar # you now have a pointless file /boot/bar
Offline
Okay thank you very much, I have removed these files.
However, my original problem does not seem to be fixed. Does anyone have any ideas in this regard?
Offline
If the version of the linux package now matches the output of `uname -a` please post an updated journal.
Offline
I updated my system and after that I could not restart my laptop normally. I was greeted by this error message when booting.
I followed your instructions and reinstalled the kernel. That worked, so I didn't end up in the emergency shell at boot time.
The system update seems to have solved my original problem so my system now boots normally.
There is no error in that output on that image, what makes you think there's an error?
I have now successfully reinstalled the kernel. In fact, my laptop is now booting successfully again.
So…
However, my original problem does not seem to be fixed. Does anyone have any ideas in this regard?
wtf. is this about?
Offline
To recap:
My original problem (which occurred after a system update) is that when I boot my laptop I get this message (which V1del says is apparently not an error message) and the boot process aborts so my laptop does not boot normally. When I chroot into my system and reinstall the kernel, I can successfully boot exactly once. If I then shut down my system again and boot it again (or reboot it), the problem described initially occurs again.
My first attempt at solving my original problem was to chroot into my system and update it again. In the process, I made the mistake of not mounting the boot partition when chrooting into my system and then updated it, which led to more problems, but with your help they are fixed so that only my original problem remains.
Last edited by archforthewin (2023-06-12 18:30:12)
Offline
If the version of the linux package now matches the output of `uname -a` please post an updated journal.
Now that my system has booted normally (after reinstalling the kernel again), the version of the linux package matches the output of "uname -a".
Here is the output of
journalctl -rb
Offline
Don't "-r", that's like top-quoting.
Also
Jun 12 20:40:31 archlaptop systemd[1244]: Reached target Current graphical user session.
that's a succesfull boot?
We'd probably need
sudo journalctl -b -1 | curl -F 'f:1=<-' ix.io
for the previous boot and don't use the power button to shutdown the system on "failure" but try ctrl+alt+del ("frenetically") or https://wiki.archlinux.org/title/Keyboa … el_(SysRq)
On a limb: the GPU shows up late, so try https://wiki.archlinux.org/title/Kernel … _KMS_start and if that doesn't help and you consistenly have boot issues, try to boot only the multi-user.target (2nd link below) and in doubt along "nomodeset", https://wiki.archlinux.org/title/Kernel_parameters
Offline
To the first question: Yes that was a successful boot.
Here is the output of
sudo journalctl -b -1 | curl -F 'f:1=<-' ix.io
which was not a successful boot.
Thank you very much for the hint regarding SysRq. I had never heard of it, but I will try to use this hint in the future.
(Possibly, since I am not a native English speaker) I am not sure if your last comment is the clear recommendation to try "Early KMS start". Are there any disadvantages to try it? Or worst case it just doesn't work and I have to undo the change to the configuration?
Offline
your last comment is the clear recommendation to try "Early KMS start"
This.
Are there any disadvantages to try it?
Nope.
Offline