You are not logged in.
Hi all,
I can't boot my Arch system anymore:
[FAILED] Failed to start Load Kernel Modules
See 'systemctl status systemd-modules-load.service' for details.
And a bit further:
[FAILED] Failed to mount /boot/efi.
See 'systemctl status boot-efi.mount' for details.
[DEPEND] Dependency failed for Local File Systems.
Then I get a message saying I am in emergency mode and asking for my root password.
Here is the result of the commands suggested:
[tan@bobby ~]$ systemctl status systemd-modules-load.service
● systemd-modules-load.service - Load Kernel Modules
Loaded: loaded (/usr/lib/systemd/system/systemd-modules-load.service; static; vendor preset: disabled)
1 Active: failed (Result: exit-code) since Sun 2018-01-28 02:10:16 NZDT; 7min ago
Docs: man:systemd-modules-load.service(8)
man:modules-load.d(5)
Process: 224 ExecStart=/usr/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE)
Main PID: 224 (code=exited, status=1/FAILURE)
Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.
[tan@bobby ~]$ systemctl status boot-efi.mount
● boot-efi.mount - /boot/efi
Loaded: loaded (/etc/fstab; generated)
Active: failed (Result: exit-code) since Sun 2018-01-28 02:10:17 NZDT; 7min ago
Where: /boot/efi
What: /dev/disk/by-uuid/F62F-E8CB
Docs: man:fstab(5)
man:systemd-fstab-generator(8)
Process: 227 ExecStart=/usr/bin/mount /dev/disk/by-uuid/F62F-E8CB /boot/efi -t vfat -o rw,realtime,fmask-0022,codepage=437,iocharset=iso-8859-1,shortname=mixed,utf8,errors=remount-ro (code=exited, status=32)
I tried restarting with systemctl status systemd-modules-load.service but it says the control process exited. For boot-efi.mount, it just says mount failed. I notice that many modules must be missing. For example I don't even see my wireless interface.
The only thing I can think of that could have caused this is that I created a package, and then installed it with pacman -U --force. I did this because I had previously installed that software from source without creating a package and wanted a "cleaner" install of it. I guess I was wrong, or I screwed up the package creation part...
Can anyone help me troubleshoot this?
Last edited by eichan (2018-01-28 17:58:38)
Offline
Which package was it you created and forced the install? Post the package build please.
All men have stood for freedom...
For freedom is the man that will turn the world upside down.
Gerrard Winstanley.
Offline
What you're seeing usually happens when you're booting the wrong kernel, such as when /boot wasn't mounted when you updated the system. Check uname -r vs pacman -Q <kernelpackage>
Offline
The package I was trying to build is libindi_3rdparty, which is a set of drivers for the INDI library (for astronomical devices such as mounts or CCD cameras). This package exists in the AUR but it's an outdated version, and the source downloaded from the author's Github page is missing a file (log4z.h) in the indi-qhy folder. This is the reason my PKGBUILD (see here) is pointing at a local source. That tarball on my local drive is just the exact same source as found on Github, just with the added log4z.h file in the right folder. makepkg and pacman -U pkg.tar.gz seemed happy with it. I uninstalled the package as root under the emergency environment provided by the failed boot, and it didn't change anything.
As for uname -a vs paman -Qi linux, both commands give me the same version number (4.14.15-1). Trying the fallback kernel on the grub menu doesn't help. Exact same problem.
Offline
I'll just add to this a photo taken of my screen as it fails its boot session. Sorry it's an image file. The emergency environment doesn't know anything about my wifi adapter (does not appear with 'ip link'), so it's a bit challenging to share some results and logs.
Offline
1. What is your bootloader? Could you please post the bootloader configuration related to you Arch install?
2. Please provide the outputs of `lsblk` and `cat /etc/fstab`.
3. Does reinstalling the kernel and/or bootloader help with the issue? (of course after uninstalling libindi_3rdparty)
Offline
1. I use grub. Here is grub.cfg.
2. lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
nvme0n1 259:0 0 477G 0 disk
├─nvme0n1p1 259:1 0 500M 0 part
├─nvme0n1p2 259:2 0 128M 0 part
├─nvme0n1p3 259:3 0 191.6G 0 part
└─nvme0n1p4 259:4 0 284.7G 0 part /
EDIT: p1 is the /boot partition, p2 is some Windows obscure partition, p3 is my Windows partition and p4 the Arch one. I don't have swap.
cat /etc/fstab
# Static information about the filesystems.
# See fstab(5) for details.
# <file system> <dir> <type> <options> <dump> <pass>
# /dev/nvme0n1p4
UUID=278106c9-fa11-4d83-b2b6-f849cf8b8a7d / ext4 rw,relatime,data=ordered 0 1
# /dev/nvme0n1p1 LABEL=ESP
UUID=F62F-E8CB /boot/efi vfat rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=iso8859-1,shortname=mixed,utf8,errors=remount-ro 0 2
3. 'pacman -S linux' exits with an error:
resolving dependencies...
looking for conflicting packages...
Packages (1) linux-4.14.15-1
Total Installed Size: 101.19 MiB
Net Upgrade Size: 0.00 MiB
checking keyring...
checking package integrity...
loading package files...
checking for file conflicts...
checking available disk space...
:: Processing package changes...
reinstalling linux...
:: Running post-transaction hooks...
(1/3) Updating linux module dependencies...
depmod: ERROR: could not open directory /lib/modules/4.14.15-1-ARCH: No such file or directory
depmod: FATAL: could not search modules: No such file or directory
(2/3) Updating linux initcpios...
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'default'
-> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux.img
==> ERROR: '/lib/modules/4.14.15-1-ARCH' is not a valid kernel module directory
==> Building image from preset: /etc/mkinitcpio.d/linux.preset: 'fallback'
-> -k /boot/vmlinuz-linux -c /etc/mkinitcpio.conf -g /boot/initramfs-linux-fallback.img -S autodetect
==> ERROR: '/lib/modules/4.14.15-1-ARCH' is not a valid kernel module directory
(3/3) Arming ConditionNeedsUpdate...
I thought grub wasn't related because it still works in the sense that it tries to run Arch and it runs Windows without a problem. And its installation is the trickiest... Do you confirm, after seeing the above, that it would be useful to try reinstalling it?
EDIT: Unsure if that's relevant but I should inform that I'm running all those commands as root, from the emergency mode.
Last edited by eichan (2018-01-28 16:53:52)
Offline
Should have been
lsblk -f
to compare the UUIDs
Online
What does `ls -l /lib` give you?
Offline
lsblk -f
NAME FSTYPE LABEL UUID MOUNTPOINT
nvme0n1
├─nvme0n1p1 vfat ESP F62F-E8CB
├─nvme0n1p2
├─nvme0n1p3 ntfs OS CEFC69F9FC69DBE9
└─nvme0n1p4 ext4 278106c9-fa11-4d83-b2b6-f849cf8b8a7d /
Offline
ls -l /lib
ls: cannot access '/lib': No such file or directory
There is a /lib64, so 'ls -l /lib64':
lrwxrwxrwx 1 root root 7 Oct 17 20:32 /lib64 -> usr/lib
And /usr/lib is full of library, here is the 'ls -l /usr/lib' in case it's relevant.
Offline
Just to be sure what is the output of
$ stat /lib /lib64 /usr/lib
Offline
stat /lib /lib64 /usr/lib
File: /lib64 -> usr/lib
Size: 7 Blocks: 0 IO Block: 4096 symbolic link
Device: 10304h/66308d Inode: 14 Links: 1
Access: (0777/lrwxrwxrwx) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2018-01-29 18:09:55.006666738 +1300
Modify: 2017-10-17 20:32:47.000000000 +1300
Change: 2018-01-05 17:19:30.660587086 +1300
Birth: -
File: /usr/lib
Size: 196608 Blocks: 392 IO Block: 4096 directory
Device: 10304h/66308d Inode: 12453953 Links: 206
Access: (0755/drwxr-xr-x) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2018-01-29 18:17:02.270828858 +1300
Modify: 2018-01-27 21:01:14.050616206 +1300
Change: 2018-01-27 21:01:14.050616206 +1300
Birth: -
Offline
Please reinstall the filesystem package that should create the symlink from /lib to /usr/lib and check the output of the command again.
Edit:
Was expecting the following output for stat /lib (assuming it was omitted)
stat: cannot stat '/lib': No such file or directory
Last edited by loqs (2018-01-28 17:26:41)
Offline
So somehow, your /usr symlink got deleted. It should point to usr/lib, same as /lib64. This is the cause of all of your problems.
Offline
Reinstalling the filesystem package solved the problem. And indeed, the No such file or directory showed was omitted by the '>' pipe.
For information, the output of the command (now run as normal user) is:
tan@bobby ~$ stat /lib /lib64 /usr/lib
File: /lib -> usr/lib
Size: 7 Blocks: 0 IO Block: 4096 symbolic link
Device: 10304h/66308d Inode: 13 Links: 1
Access: (0777/lrwxrwxrwx) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2018-01-29 19:53:11.046666744 +1300
Modify: 2017-10-17 20:32:47.000000000 +1300
Change: 2018-01-29 19:52:07.698039494 +1300
Birth: -
File: /lib64 -> usr/lib
Size: 7 Blocks: 0 IO Block: 4096 symbolic link
Device: 10304h/66308d Inode: 14 Links: 1
Access: (0777/lrwxrwxrwx) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2018-01-29 19:52:07.818039500 +1300
Modify: 2017-10-17 20:32:47.000000000 +1300
Change: 2018-01-29 19:52:07.698039494 +1300
Birth: -
File: /usr/lib
Size: 196608 Blocks: 392 IO Block: 4096 directory
Device: 10304h/66308d Inode: 12453953 Links: 206
Access: (0755/drwxr-xr-x) Uid: ( 0/ root) Gid: ( 0/ root)
Access: 2018-01-29 19:52:07.711372828 +1300
Modify: 2018-01-29 19:52:07.698039494 +1300
Change: 2018-01-29 19:52:07.698039494 +1300
Birth: -
So do you know how this could have happened? Is my PKGBUILD completely wrong? Maybe by pacman -R libindi_3rdparty is somehow deleted the /usr simlink?
Many thanks to all.
Offline
If you used --force to install it, it could have. If it contained files in /lib, using --force would have overwritten the symlink with a dir and everything went to hell from there.
Offline
Aaaah OK it all makes sense now. Lesson learned. Thanks again.
Offline