[SOLVED] Cannot boot - stuck on dev/sda after grub

Hi everyone
I posted the below yesterday on the announcement thread related to the latest update:

Anyone faced issues with "failed to start light display manager"? I got this from running journalctl -p err -b -1. In mhwd -li i have just video-hybrid-intel-nouveau-bumblebee installed. I reinstalled it too when I saw this issue. I do have lightdm-gtk-greeter installed. Should I just reboot and see what happens? If i do sudo mhwd-gpu --setgl it says invalid argument setgl. If I do setxorg it does nothing.

Today I started the machine and i got this message

I know that light display manager has to do with the graphics card. So yesterday I removed then installed the ....nvidia-bumblebee driver and when that didn't work i removed it and installed .....nouveau-bumblebee driver instead but the error about light display manager remained the same. I'm not sure what to do really. I have only one driver there's nothing conflicting. The update from yesterday was eventually done successfully without errors by updating keyring then manjaro system and then removing mhwd and reinstalling it. I did the pacman command which refreshes the database and it refreshed them all with no errors or warnings. I'm using the Linux 44 kernel LTS. Thanks in advance!

Look at this guy - maybe just as the worst incidence one of your usb devices decided to die:
https://paulphilippov.com/articles/how-to-fix-device-not-accepting-address-error

Just unplug everything USB?

Thanks. This solved the additional usb rows but not the boot problem. I noticed though a message when removing the driver : primus to be removed before its dependency bumblebee. Here's the full message
https://pastebin.com/evnypuvr

And when i tried to install the free driver it was the same thing even though it said successfully installed.
https://pastebin.com/yjdzya2f

I tried to remove lib32-primus (read in another thread) but it said could not find or read directory lib32-primus/var/lib/pacman

Please look at this thread

From here downwards:

and also please check if in /etc/bumblebee there is a .conf.pacnew file - if so save the original bumblebee.conf and use the pacnew version - as it has undergone major change.

I hope that helps - lacking switchable graphics myself I have not done the fix myself though

1 Like

I already am up to date. I did the steps again though and after syyu it said there is nothing to do. The output from journalctl changed though

Apr 04 00:15:16 ruffy-pc ntpd[4388]: bind(21) AF_INET6 fe80::1464:a8fa:86d4:291%3#123 flags 0x11 failed: Cannot assign requested address
Apr 04 00:15:16 ruffy-pc ntpd[4388]: unable to create socket on wlp3s0 (5) for fe80::1464:a8fa:86d4:291%3#123
Apr 04 00:25:30 ruffy-pc bumblebeed[665]: Could not open /proc/acpi/bbswitch: No such file or directory
Apr 04 00:25:30 ruffy-pc bumblebeed[665]: Could not enable discrete graphics card

I found the file it's here:
/etc/lightdm/lightdm.conf.pacnew

And i found the old bumblebee files

/etc/bumblebee/bumblebee.conf
/etc/bumblebee/bumblebee.conf.pacsave
/etc/bumblebee/bumblebee.conf.pacsave.1
/etc/bumblebee/bumblebee.conf.pacsave.2
/etc/bumblebee/bumblebee.conf.pacsave.3
/etc/modprobe.d/bumblebee.conf

It looks that they are already saved. I don't understand how i can tell it to use the new conf.pacnew instead of the old files. Should I delete the first 5 files from the output?

  1. it always uses the config file that is nameofthepackage.conf without any further endings. So bumblebee.conf.pacsave are actually old and older files that got replaced by pacman with newer ones already.
    (with ls -lh you will see the timestamps so you can tell which one is the newest)
  • but try the lightdm stuff first (2) - than I would try and reinstall bumblebee if it was my machine.
  1. for /etc/lightdm/lightdm.conf.pacnew - there will be a lightdm.conf as well - move (mv) that to lightdm.conf.old and than cp lightdm.conf.pacnew to lightdm.conf that way that file will be used.

Can not promise, but probably that solves it for you.

nope that didn't work. But I reinstalled video-hybrid-intel-nvidia-bumblebee. I got the primus bumblebee dependency message:
www.pastebin.com/prunqrq0
But install was successful and I was able to boot normally. So that's good :slight_smile:
I also now updated to the new recommended kernel 4.9.20-1

I'm a bit worried cos I did the journalctl and got several errors, not sure if they are serious or how much:
[ruffy@ruffy-pc ~]$ journalctl -p err -b -1
-- Logs begin at Fri 2017-01-06 22:07:57 EET, end at Tue 2017-04-04 23:24:12 EEST. --
Apr 04 22:52:51 ruffy-pc ntpd[1614]: bind(21) AF_INET6 fe80::1464:a8fa:86d4:291%3#123 flags 0x11 failed: Cannot assign requested addre
Apr 04 22:52:51 ruffy-pc ntpd[1614]: unable to create socket on wlp3s0 (5) for fe80::1464:a8fa:86d4:291%3#123
Apr 04 22:59:47 ruffy-pc kernel: usb 2-8: can't restore configuration #1 (error=-71)
Apr 04 22:59:49 ruffy-pc kernel: usb 2-8: can't set config #1, error -71
Apr 04 23:22:33 ruffy-pc kernel: watchdog watchdog0: watchdog did not stop!

Thank you for your help

That - most likely is because there is no IPv6 on your network so ntp (network time protocol) can not bind to a ipv6 address - do not pin me down on that though :sweat:

Some of your USB connected hardware/peripherals is acting up - needs further invesigation (same errors you saw beginning the threat actually)

Some watchdog timer did not stop - well that does not seem to serious to me - if the PC does not freeze or reboot out of a sudden - that watchdoog probably just not beeing used - or part of the USB problem - maybe

So if your system is running - its not serious =D

When curious - start a new thread (maybe in a view days when all the problems with the last update are slowely solved for everyone) for it and give it the appropriate tags =)

PS: now that your system works - plz mark that thread as solved <3

I changed the mouse and today it only said this:

[ruffy@ruffy-pc ~]$ journalctl -p err -b -1
-- Logs begin at Fri 2017-01-06 22:07:57 EET, end at Wed 2017-04-05 19:23:08 EES
Apr 04 23:23:11 ruffy-pc kernel: platform wdat_wdt: failed to claim resource 4
Apr 04 23:23:12 ruffy-pc kernel: ACPI: watchdog: Failed to create platform devic
lines 1-3/3 (END)...skipping...
-- Logs begin at Fri 2017-01-06 22:07:57 EET, end at Wed 2017-04-05 19:23:08 EEST. --
Apr 04 23:23:11 ruffy-pc kernel: platform wdat_wdt: failed to claim resource 4
Apr 04 23:23:12 ruffy-pc kernel: ACPI: watchdog: Failed to create platform device

I will check it out later on. Huge thanks for your help!!:relaxed:

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.

Forum kindly sponsored by