ERROR: root device mounted succesfully, but /sbin/init does not exists

Support
1

Hi,
I saw that a stable update was launched and I updated the system today. I recieved a message that there were errors during the update and I decided to reboot.
Afterwards, an ERROR message and I ended in the rootfs prompt.
I read the posts related with my problem and found that I should run
mkinitcpio -P
The question is the following:
I have Arcolinux installed in a separate disk and I can access to the Manjaro root folder as root through Thunar and to the etc folder where mkinitcpio is available.

I also ckeck that the file init in the sbin folder is present.

I wonder whether I can run the following terminal command inside the root folder:
sudo /etc/mkinitcpio -P

I tried to run this command under the root prompt typing sudo su - at the rootfs prompt but I did not run.

This is the output of the disks.

lsblk -af
NAME       FSTYPE FSVER LABEL        UUID                                 FSAVAIL FSUSE% MOUNTPOINTS
sda                                                                                      
├─sda1     vfat   FAT32              3686-B6F3                                           
├─sda2     swap   1     swap         d7080cd4-6338-477d-a36e-2509cd444e9c                
├─sda3     ext4   1.0                1f905f6b-acee-4acd-96d3-9501d3b47e6a                
└─sda4     ext4   1.0   Manjaro_home 5e31ce64-3f4a-4d63-99f8-806d9d69d9a7   90.7G    19% /run/media/jochoa/Manjaro_home
mmcblk0                                                                                  
├─mmcblk0p1
│          vfat   FAT32              158D-0B61                             299.1M     0% /boot/efi
├─mmcblk0p2
│          ext4   1.0                dd313c81-a658-4ed5-9166-3525053f7723    1.6G    89% /
└─mmcblk0p3
           swap   1     swap         089f98af-eb9f-43b1-8300-d68a93bd4f41                [SWAP]
mmcblk0boot0
                                                                                         
mmcblk0boot1

This is the output of blkid

 blkid 
/dev/sda4: LABEL="Manjaro_home" UUID="5e31ce64-3f4a-4d63-99f8-806d9d69d9a7" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="8bea6853-5ba2-4212-bffb-ec30f6243d78"
/dev/sda2: LABEL="swap" UUID="d7080cd4-6338-477d-a36e-2509cd444e9c" TYPE="swap" PARTUUID="53175558-e0e7-44e4-8710-7153756bf190"
/dev/sda3: UUID="1f905f6b-acee-4acd-96d3-9501d3b47e6a" BLOCK_SIZE="4096" TYPE="ext4" PARTUUID="01ac760c-9ae9-4902-993d-52686f81bfff"
/dev/sda1: UUID="3686-B6F3" BLOCK_SIZE="512" TYPE="vfat" PARTUUID="508cba71-9a45-485f-8948-b2fd1214ef78"

This is the output of pacman.log

[2024-01-02T11:58:56-0300] [PACMAN] Running 'pacman -Syyu'
[2024-01-02T11:58:56-0300] [PACMAN] synchronizing package lists
[2024-01-02T12:00:15-0300] [ALPM] transaction started
[2024-01-02T12:00:15-0300] [ALPM] upgraded archlinux-keyring (20231207-1 -> 20231222-1)
[2024-01-02T12:00:15-0300] [ALPM-SCRIPTLET] ==> Appending keys from archlinux.gpg...
[2024-01-02T12:00:24-0300] [ALPM-SCRIPTLET] ==> Updating trust database...
[2024-01-02T12:00:24-0300] [ALPM-SCRIPTLET] gpg: next trustdb check due at 2024-01-21
[2024-01-02T12:00:24-0300] [ALPM] transaction completed
[2024-01-02T12:00:24-0300] [ALPM] running '30-systemd-daemon-reload.hook'...
[2024-01-02T12:00:26-0300] [ALPM] running '30-systemd-update.hook'...
[2024-01-02T12:00:26-0300] [PACMAN] starting full system upgrade
[2024-01-02T12:04:13-0300] [ALPM] transaction started
[2024-01-02T12:04:13-0300] [ALPM] upgraded tzdata (2023c-2 -> 2023d-1)
[2024-01-02T12:04:13-0300] [ALPM] upgraded fontconfig (2:2.14.2-1 -> 2:2.15.0-1)
[2024-01-02T12:04:13-0300] [ALPM-SCRIPTLET] Rebuilding fontconfig cache...
[2024-01-02T12:04:56-0300] [ALPM] upgraded alacritty (0.12.3-1 -> 0.13.0-3)
[2024-01-02T12:04:56-0300] [ALPM] upgraded c-ares (1.23.0-1 -> 1.24.0-1)
[2024-01-02T12:04:56-0300] [ALPM] upgraded cdparanoia (10.2-8 -> 10.2-9)
[2024-01-02T12:04:56-0300] [ALPM] upgraded libdrm (2.4.118-1 -> 2.4.119-1)
[2024-01-02T12:04:56-0300] [ALPM] upgraded krb5 (1.20.1-2 -> 1.21.2-2)
[2024-01-02T12:04:56-0300] [ALPM] upgraded systemd-libs (254.7-1 -> 254.8-1)
[2024-01-02T12:04:56-0300] [ALPM] upgraded coin (4.0.0.04c190-4 -> 4.0.2-2)
[2024-01-02T12:04:56-0300] [ALPM] upgraded libseccomp (2.5.4-2 -> 2.5.5-1)
[2024-01-02T12:04:57-0300] [ALPM] upgraded ctags (1:6.0.0-1 -> 1:6.1.0-1)
[2024-01-02T12:04:57-0300] [ALPM] upgraded dhcpcd (10.0.5-1 -> 10.0.6-1)
[2024-01-02T12:04:57-0300] [ALPM] upgraded exempi (2.6.4-1 -> 2.6.5-1)
[2024-01-02T12:04:57-0300] [ALPM] error: could not extract /usr/lib/security/pam_systemd_home.so (Zstd decompression failed: Data corruption detected)

Any suggestion will be welcomed!
Regards

2

Binaries do not reside in /etc. Where you got that idea?

Anyway, you need to use a live usb image, chroot into manjaro install and run update again - you might want to delete /var/cache/pacman/pkg/systemd-* packages so it re-downloads them again. You might also have missing kernels, so you might want to install whichever you were using again.

3

Rebooting was a bad idea.

And, even if it could run, mkinitcpio wont fix this.

You have a borked upgrade that only begun to be applied when it failed, which would be a partial-upgrade state in and of itself.

But it also failed trying to extract the upgrade for systemd (in relation to pam) … which would also be a problem.

What you need to do is … ah, @zbe beat me to it.

Note: If it is truly that single error/file then the package would be systemd … you should not need to remove/redownload all other systemd-* cached packages.

Also … theres some guides around … like this one:

1 Like
4

Hi @zbe and @cscs ,

Thank you for your answers,

5

This topic was automatically closed 3 hours after the last reply. New replies are no longer allowed.