Booting into some kind of log instead of GUI

Support
21

To me, this is
dmesg
output.

It looks exactly like what you get when you run:
sudo dmesg

or some variation on that command
in a terminal or on a TTY.

I can not reproduce this behaviour
not even when I purposely disable my display manager (sddm in KDE)
and have no idea how to debug this
when you are unable to switch to at least a TTY and run commands there to investigate

Using a live medium to boot and then chroot into the system and then examining the logs for any error or failure message there
would be my strategy.

1 Like
22

I looked online It is indeed exactly like the output from
dmesg
But nowhere online i see someone getting it on boot :grin:
Will chroot into my manjaro installation from a live usb boot
And see what i can do…
will update what i find out when i do so, Thanks alot!

23

No, it’s the output from journald, which by default goes to tty12 ─ if properly configured, because that didn’t use to be set up by default in Manjaro.

Somehow, pressing that Fn+F3 combination locked the OPs laptop into tty12, which means it must have messed with his boot-up files, and particularly, with the systemd configuration.

24

Heres another bit of text hope it helps in identifying the nature of the problem.

25

I suspected as much - but I could not reproduce.
I only have Manjaro in a VM
and if I go to tty12 - there is nothing there
not by default anyway
and when I run
journalctl -k
(or any other switch I tried)
the output format is different
not what is shown here.
:man_shrugging:

26

Okay, that is the output of dmesg, but what you showed earlier was not.

Is this a virtual machine by any chance?

27

… and, at the top few lines, there is evidence of a crashed program
so it seems to me
but too few lines to say which process

28

No i have it installed in a seperate partition from my windows installation.
The previous image was at the same screen as the second image the only difference is i plugged in a flash drive which logged on the screen.

29

Can you boot up from the install medium in live mode?

30

Yes planning to do that will take bit of time till i download manjaro and put it on a flash…
Will update when i do so

31

I have to add something i just remembered i did this morning which is deleting pacman cache (i always do that no problem)
But what i did today is i also deleted the contents of ~/.cache
I dont know if this relates to the problem because the OS was working fine for a better of 4 hours before the error appeared.

32

No, it is perfectly fine to delete ~/.cache. That cannot have caused your problem.

1 Like
33

Before i go on and chroot i Tried to edit the boot options by adding init=/bin/bash and make the system boot into bash and it worked! Anything i can do here to get useful info or probably timeshift or i should go ahead and chroot?

34

not sure
but you could try to start sddm from there
systemctl start sddm
or just
startx
and observe the result?

and/or examine the logs from where you are now
just like as you would when in chroot

chroot by itself doesn’t do anything special

35 
systemctl start sddm

Gets this message “System has not been booted with systemd as init system (Pid 1). Cant operate.
Failed to connect to bus: Host is down”

startx

Looks like it attempts to start the display manager but it fails and i get some error messages

36

Can you please share what command i should use?

37

that’s because
“System has not been booted with systemd as init system (Pid 1)”
as you used
init=/bin/bash

… I think …

try just appending a “1” to the boot options
instead of
init=/bin/bash

38

I tried it now it sent me back to the first screen with the logs

39

It should have sent you into single user/rescue mode
and asked for the admin password to proceed …

apparently not

an admin password is probably not set anyway - the sudo password will not work as that is the user’s password

so then it’s probably better to chroot and examine the log from there
journalctl -r
or something like that, which lets you find the latest error messages most easily
I’m not very familiar with the tool - use
man journalctl
or maybe someone else may know a quicker/better way to filter the logs
because
I don’t :man_shrugging:

ps:
Just to be sure
when booting, at the grub prompt, press “E”
and add a space and then a 1 to the end of the kernel line
just after the “ro”
which is last …

1 Like
40

I added " 1" at the end of the line but theres more text between the ro and where i put the 1 i could go back and take a picture if it is of help