SDDM crash and random X crashes since upgrade

Storm · 2017-01-13 00:27:42

I just dud a full sysupgrade yesterday, and ever since I'm in hell. I have two major issues and I suspect they are related so I outline both here.

Issue #1

After upgrade SDDM simply crashes after boot. A friend of mine and a channel op in #archlinux both tried to help me to no avail. The latter, Namarrgon asked a shitton of info from me and made me try dozens of things but nothing helped.
- There is nothing relevant in Xorg.0.log
- Journal shows the crash (Auth: sddm-helper exited with 7) but not what causes it. Could not find an explanation of what error number 7 means.
- Tried reinstalling nvidia driver and X
- Tried removing ~/.cache and /var/lib/sddm/.cache
- Tried downgrading updated packages to pre-update state
- Tried generating a fresh xorg.conf

journalctl -b: https://ptpb.pw/wQGy.txt
pacman log: https://ptpb.pw/Yz4h
Installed nVidia packages: https://ptpb.pw/53r_
pacman -Qm: https://ptpb.pw/5ypq

After seven and a half hours of no success in getting any closer Namarrgon told me to boot into multi-user.target and try starting kde manually through ~/.xinitrc.

This caused KDe to start up normally, apparently without problem, simply bypassing SDDM altogether. So now I'm logging in from terminal and starting kde via "exec startic" from my .bash-profile. I can use my system now, but the SDDM issue is still a mystery.

Issue #2

While I can get into KDE now, I have a new issue I didn't have before, and due to it's nature In suspect it is connected with the first one.

Every now and then (happened to me three times with 1-1.5 hours delays) X randomly crashes - after 1-2 sec of black screen I'm back on tty1, logged out, just like the state after boot. I was painting in Krita, and had no crash since I stopped, so heavy use of graphics may be related.

- CPU, GPU and MoBo temp is normal (30-35). Could not measure after crash.
- Nothing in Xorg logs. It goes on to periodically report on status then it ends abruptly with nothing related to the crash.
- The journal only shows X already being dead and stack traces of programs dying because of X is dead, nothing about why or how X died.

journalctl -b snippets:
http://pastebin.com/kYErgyc2
http://pastebin.com/e5jWdbAp
http://pastebin.com/FwArBP93

Xorg.0.log: http://pastebin.com/7qkR365z
Xorg.1.log: http://pastebin.com/MYmf15t7
Xorg.2.log: http://pastebin.com/1RTHaJ80

Note: Xorg.1 and 2 logs because the system did not crash, but I started a new X session after the previous dying.

mrunion · 2017-01-13 15:04:09

So is your system currently fully up to date? Are you ignoring any packages? Do you have any of the infinality font stuff installed? What is the output of uname -a?

PootisBird · 2017-01-13 20:01:21

My sddm cried about kdeinit5 being broken. I removed infinality and not it works fine.
Thanks for the hint @mrunion.

Storm · 2017-01-14 11:27:26

mrunion wrote:

So is your system currently fully up to date? Are you ignoring any packages? Do you have any of the infinality font stuff installed? What is the output of uname -a?

- It was up to date when SDDM died and when I had X crashed*
- Not ignoring any package.
- No infinality, just freetype2 2.7.1-1
- Linux Camelot 4.8.13-1-ARCH #1 SMP PREEMPT Fri Dec 9 07:24:34 CET 2016 x86_64 GNU/Linux

Still having crashes but only when I work in Krita, and after about an hour +-30 min spent working in it. If I don't run Krita or anything heavy no crash so far.

* Today there are new updates, so doing Syu; I'll report back if the X crash is still occurring afterwards.

Right now my priority is Issue #2, I can live without SDDM but I can't work of my X crashes every hour (I'm a freelancer artist, I work from home).

Smjert · 2017-01-14 12:01:35

Hello, i had (September 2016) and have now the same dreaded bug as Storm: black screen at boot, sddm-helper exited with 7 in the logs, sddm-greeter crashing with a stack trace mentioning libGLdispatch.

Back then it happened after an update that updated sddm and nvidia, and it was like 1 month i wasn't updating Arch.
I've tried the usual things, deleting ~/.cache, /var/lib/sddm/.cache, deleting sddm.conf and making sddm recreate it, reinstall sddm multiple times even after having deleted any trace of sddm files and folders, reinstall nvidia.
I've also tried to compile sddm in debug mode to try to understand something more but had not much success...
Also tried downgrading to a two months before set of packages thanks to the archive repository, when i was sure it was working... but it did had the same issues O.o.
The only way it would work it was with nouveau driver instead of official Nvidia ones...

After all this and a couple of days of madness, it magically worked again after something updated (but i don't recall what it was.. pretty sure it wasn't sddm or nvidia drivers).

And now it's the same, obviously i'm not trying again all the above... but i wonder why this is happening again..

Last edited by Smjert (2017-01-14 12:02:25)

Storm · 2017-01-14 13:05:13

Here is a step-by-step of the workaround bypassing SDDM I'm using for SDDM if anyone else need it:

1; Boot into multi-user.target instead of the graphical. This can be done by modifying the boot loader parameters or by running this (will be effective next reboot):

systemctl set-default multi-user.target

Next time you reboot you will be greeted with a terminal, X will not be started automatically.

2; If you don't have xinit, install it from the extra repository:

pacman -S xorg-xinit

If you don't have ~/.xinitrc, create it by copying the default:

cp /etc/X11/xinit/xinitrc ~/.xinitrc

3; If you copied the default, disable these lines by deleting or commenting them:

#twm &
#xclock -geometry 50x50-1+1 &
#xterm -geometry 80x50+494+51 &
#xterm -geometry 80x20+494-0 &
#exec xterm -geometry 80x66+0+0 -name login

4; Append exec startkde to the end of your .xinitrc

At this point my .xinitrc looks like this:

#!/bin/sh

userresources=$HOME/.Xresources
usermodmap=$HOME/.Xmodmap
sysresources=/etc/X11/xinit/.Xresources
sysmodmap=/etc/X11/xinit/.Xmodmap

# merge in defaults and keymaps

if [ -f $sysresources ]; then







    xrdb -merge $sysresources

fi

if [ -f $sysmodmap ]; then
    xmodmap $sysmodmap
fi

if [ -f "$userresources" ]; then







    xrdb -merge "$userresources"

fi

if [ -f "$usermodmap" ]; then
    xmodmap "$usermodmap"
fi

# start some nice programs

if [ -d /etc/X11/xinit/xinitrc.d ] ; then
 for f in /etc/X11/xinit/xinitrc.d/?*.sh ; do
  [ -x "$f" ] && . "$f"
 done
 unset f
fi

#twm &
#xclock -geometry 50x50-1+1 &
#xterm -geometry 80x50+494+51 &
#xterm -geometry 80x20+494-0 &
#exec xterm -geometry 80x66+0+0 -name login

exec starkde

5; OPTIONAL: To start Xorg automatically after you log in on TTY1, add this to your ~/.bash_profile

# If I just logged in on tty1, start KDE
if [ -z "$DISPLAY" ] && [ -n "$XDG_VTNR" ] && [ "$XDG_VTNR" -eq 1 ]; then
  exec startx
fi

Otherwise after logging in type startx to start Xorg manually.

At this point kde should start automatically via .xinitrc

Kudos to Namarrgon who showed me this

Last edited by Storm (2017-01-14 13:15:48)

WorMzy · 2017-01-14 13:32:11

Please post full journals, not just snippits.

pacman -Qm: https://ptpb.pw/5ypq

You still have remnants of KDE4 installed, which could be causing problems.
https://www.archlinux.org/news/dropping-plasma-4/

Clean out your old packages. Aside from the KDE4 stuff, 'pacman -Qtd' may help to identify other packages that you no longer need.

Xorg.0.log wrote:

[  3848.273] (==) Using config file: "/etc/X11/xorg.conf"

Why do you have a xorg.conf? What is in it?

Storm · 2017-01-14 14:37:37

WorMzy wrote:

Please post full journals, not just snippits.

https://www.dropbox.com/s/waj1q3s9hdox6 … l.txt?dl=0
X stopping at line 2753.

WorMzy wrote:

You still have remnants of KDE4 installed, which could be causing problems. [...] Clean out your old packages. Aside from the KDE4 stuff, 'pacman -Qtd' may help to identify other packages that you no longer need.

Removed:

katepart4
kdebase-lib
kdemultimedia-mplayerthumbs
kdepimlibs4
kfilemetadata4
libbaloo4
libkactivities4
libkdcraw4
libkexiv2_4
libkface4
libkgeomap4
libkipi4
libkscreen4
poppler-qt4
strigi

The rest I'll have to go through one by one to find out what do I need. Many of those were installed as build dependencies:

$ pacman -Qtd
autoconf2.13 2.13-4
bzr 2.7.0-1
cargo 0.15.0-1
elfutils 0.168-1
extra-cmake-modules 5.29.0-2
fontforge 20161012-3
ftjam 2.5.3rc2-4
gamin 0.1.10-8
gnome-common 3.18.0-1
gobject-introspection 1.50.0+1+gb8d92b0-2
gpgmepp 16.08.3-1
imake 1.0.7-1
kdesignerplugin 5.29.0-1
lib32-libdaemon 0.14-5
lib32-libpng12 1.2.57-1
lib32-libxkbcommon 0.7.0-1
lib32-portaudio 190600_20161030-1
libakonadi-qt4 1.13.0-12
libappindicator-gtk3 12.10.0-12
libappindicator-sharp 12.10.0-12
libconfig 1.5-3
libftdi-compat 0.20-3
libirman 0.5.2-1
libpgf 7.15.32-1
libpng12 1.2.57-1
libunity 7.1.4-2
ninja 1.7.2-1
pth 2.0.7-5
qimageblitz 0.0.6-5
qjson 0.8.1-5
qt5-translations 5.7.1-1
ragel 6.9-3
rest 0.8.0-1
telegramqml 0.9.2-2
vala 0.34.4-1
w3m 0.5.3.git20170102-1
x264 2:148.20160920-1
xapian-core 1:1.4.2-1
xf86-input-evdev 2.10.4-2
xorg-server-xvfb 1.19.1-1
yasm 1.3.0-1

WorMzy wrote:

Why do you have a xorg.conf? What is in it?

There should be one by default, no?

# nvidia-xconfig: X configuration file generated by nvidia-xconfig
# nvidia-xconfig:  version 352.41  (buildmeister@swio-display-x64-rhel04-11)  Sat Aug 22 00:01:10 PDT 2015


Section "ServerLayout"
    Identifier     "Layout0"
    Screen      0  "Screen0" 0 0
    InputDevice    "Keyboard0" "CoreKeyboard"
    InputDevice    "Mouse0" "CorePointer"
EndSection

Section "Files"
EndSection

Section "InputDevice"

    # generated from default
    Identifier     "Mouse0"
    Driver         "mouse"
    Option         "Protocol" "auto"
    Option         "Device" "/dev/psaux"
    Option         "Emulate3Buttons" "no"
    Option         "ZAxisMapping" "4 5"
EndSection

Section "InputDevice"

    # generated from default
    Identifier     "Keyboard0"
    Driver         "kbd"
EndSection

Section "Monitor"
    Identifier     "Monitor0"
    VendorName     "Unknown"
    ModelName      "Unknown"
    HorizSync       28.0 - 33.0
    VertRefresh     43.0 - 72.0
    Option         "DPMS"
EndSection

Section "Device"
    Identifier     "Device0"
    Driver         "nvidia"
    VendorName     "NVIDIA Corporation"
EndSection

Section "Screen"
    Identifier     "Screen0"
    Device         "Device0"
    Monitor        "Monitor0"
    DefaultDepth    24
    SubSection     "Display"
        Depth       24
    EndSubSection
EndSection

Last edited by Storm (2017-01-14 14:40:50)

Storm · 2017-01-14 16:09:05

Just had another crash.

This time got a backtrace in Xorg.0.log!!!
https://www.dropbox.com/s/dhvlq6qbqjdec … 0.log?dl=0

Journal:
https://www.dropbox.com/s/wjxu90ieeqlxe … 2.txt?dl=0

ugjka · 2017-01-14 16:14:08

~~Do you have the entire Xorg group installed with every possible driver that is there?~~
What's the output of pacman -Qs xf86-video

Last edited by ugjka (2017-01-14 16:17:43)

Storm · 2017-01-14 18:44:10

ugjka wrote:

What's the output of pacman -Qs xf86-video

local/xf86-video-fbdev 0.4.4-6 (xorg-drivers)                                                                                                                                                                  
    X.org framebuffer video driver                                                                                                                                                                             
local/xf86-video-vesa 2.3.4-3 (xorg-drivers xorg)                                                                                                                                                              
    X.org vesa video driver

NOTE: I'm using the proprietary nVidia driver. A few other people mentioned having similar crashes in the past, all of them also uses the nVidia driver.

Edit: In case you would like to know what nvidia related packages are installed:

$ pacman -Qs nvidia
local/cuda 8.0.44-3
    NVIDIA's GPU programming toolkit
local/lib32-libvdpau 1.1.1-2
    Nvidia VDPAU library
local/lib32-nvidia-libgl 375.26-2
    NVIDIA drivers libraries symlinks (32-bit)
local/lib32-nvidia-utils 375.26-2
    NVIDIA drivers utilities (32-bit)
local/libvdpau 1.1.1-2
    Nvidia VDPAU library
local/libxnvctrl 375.26-1
    NVIDIA NV-CONTROL X extension
local/nvidia 375.26-1
    NVIDIA drivers for linux
local/nvidia-libgl 375.26-2
    NVIDIA drivers libraries symlinks
local/nvidia-settings 375.26-1
    Tool for configuring the NVIDIA graphics driver
local/nvidia-utils 375.26-2
    NVIDIA drivers utilities
local/opencl-nvidia 375.26-2
    OpenCL implemention for NVIDIA

Last edited by Storm (2017-01-14 18:46:32)

Storm · 2017-01-15 13:16:45

Another crash.

This time using the stable appimage of Krita to see if that makes any difference.

Journal - crash at 14:04 https://www.dropbox.com/s/noriwyidx65r8 … 3.txt?dl=0
Nothing in Xorg log.

ugjka · 2017-01-15 13:24:01

man coredumpctl

post actual core dumps if you have any

Last edited by ugjka (2017-01-15 13:24:14)

oeut1341 · 2017-01-15 17:04:13

Having same issue here, Xorg crashing about twice a day. Started happening after last KDE update (5.8.5).

Linux (...) 4.8.13-1-ARCH #1 SMP PREEMPT Fri Dec 9 07:24:34 GMT 2016 x86_64 GNU/Linux

Fully updated, using nvidia drivers (375.26) too. Maybe that's related?

Fresh core dump (~30 min ago): https://mega.nz/#!DdwjARCR!HctYG4Kwp7Kb … waeAmpJo78

Last edited by oeut1341 (2017-01-15 17:05:38)

oeut1341 · 2017-01-16 15:22:20

Crashed another couple of times today..... backtrace shows up in Xorg log only rarely:

[  5919.052] (EE)
[  5919.052] (EE) Backtrace:
[  5919.052] (EE) 0: /usr/lib/xorg-server/Xorg (OsLookupColor+0x139) [0x59b7e9]
[  5919.052] (EE) 1: /usr/lib/libpthread.so.0 (__restore_rt+0x0) [0x7faf8abbb07f]
[  5919.052] (EE) 2: /usr/lib/xorg-server/Xorg (WriteToClient+0x21c) [0x59b00c]
[  5919.053] (EE) 3: /usr/lib/xorg-server/Xorg (DeleteAllWindowProperties+0x625) [0x4570f5]
[  5919.053] (EE) 4: /usr/lib/xorg-server/Xorg (SendErrorToClient+0x405) [0x436955]
[  5919.053] (EE) 5: /usr/lib/xorg-server/Xorg (InitFonts+0x428) [0x43a898]
[  5919.053] (EE) 6: /usr/lib/libc.so.6 (__libc_start_main+0xf1) [0x7faf8a82c291]
[  5919.053] (EE) 7: /usr/lib/xorg-server/Xorg (_start+0x2a) [0x42453a]
[  5919.053] (EE) 8: ? (?+0x2a) [0x2a]
[  5919.053] (EE)
[  5919.053] (EE) Segmentation fault at address 0xffffffff
[  5919.053] (EE)
Fatal server error:
[  5919.053] (EE) Caught signal 11 (Segmentation fault). Server aborting
[  5919.053] (EE)
[  5919.053] (EE)
Please consult the The X.Org Foundation support
         at http://wiki.x.org
 for help.
[  5919.053] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[  5919.053] (EE)
[  5919.945] (EE) Server terminated with error (1). Closing log file.

Storm · 2017-01-18 10:30:48

oeut1341 wrote:

Fresh core dump (~30 min ago): https://mega.nz/#!DdwjARCR!HctYG4Kwp7Kb … waeAmpJo78

For me that's an empty archive

Storm · 2017-01-18 10:38:05

Curedump for latest crash: https://www.dropbox.com/s/ne4cvdkrjn0cm05/corecump?dl=0

Info:

           PID: 24877 (kglobalaccel5)
           UID: 1000 (storm)
           GID: 1000 (storm)
        Signal: 6 (ABRT)
     Timestamp: Wed 2017-01-18 11:18:23 CET (16min ago)
  Command Line: /usr/bin/kglobalaccel5
    Executable: /usr/bin/kglobalaccel5
 Control Group: /user.slice/user-1000.slice/user@1000.service/dbus.service
          Unit: user@1000.service
     User Unit: dbus.service
         Slice: user-1000.slice
     Owner UID: 1000 (storm)
       Boot ID: 4088eea7570d47c9bdbc4848b9ac3de4
    Machine ID: bb660b96947649e4b7745b1301f58f1c
      Hostname: Camelot
       Storage: /var/lib/systemd/coredump/core.kglobalaccel5.1000.4088eea7570d47c9bdbc4848b9ac3de4.24877.1484734703000000000000.lz4
       Message: Process 24877 (kglobalaccel5) of user 1000 dumped core.
                
                Stack trace of thread 24877:
                #0  0x00007f3ed6aff04f raise (libc.so.6)
                #1  0x00007f3ed6b0047a abort (libc.so.6)
                #2  0x00007f3ed72854f1 _ZNK14QMessageLogger5fatalEPKcz (libQt5Core.so.5)
                #3  0x00007f3ece4b08fe _ZN14QXcbConnectionC2EP19QXcbNativeInterfacebjPKc (libQt5XcbQpa.so.5)
                #4  0x00007f3ece4b3b0e _ZN15QXcbIntegrationC1ERK11QStringListRiPPc (libQt5XcbQpa.so.5)
                #5  0x00007f3ece77430d n/a (libqxcb.so)
                #6  0x00007f3ed79aa33d _ZN27QPlatformIntegrationFactory6createERK7QStringRK11QStringListRiPPcS2_ (libQt5Gui.so.5)
                #7  0x00007f3ed79b868b _ZN22QGuiApplicationPrivate25createPlatformIntegrationEv (libQt5Gui.so.5)
                #8  0x00007f3ed79b918d _ZN22QGuiApplicationPrivate21createEventDispatcherEv (libQt5Gui.so.5)
                #9  0x00007f3ed748002f _ZN23QCoreApplicationPrivate4initEv (libQt5Core.so.5)
                #10 0x00007f3ed79babaf _ZN22QGuiApplicationPrivate4initEv (libQt5Gui.so.5)
                #11 0x00007f3ed79bb814 _ZN15QGuiApplicationC2ERiPPci (libQt5Gui.so.5)
                #12 0x0000000000401433 main (kglobalaccel5)
                #13 0x00007f3ed6aec291 __libc_start_main (libc.so.6)
                #14 0x000000000040178a _start (kglobalaccel5)

Storm · 2017-01-18 10:50:05

Another threads possibly related:
https://bbs.archlinux.org/viewtopic.php?pid=1684749
https://bbs.archlinux.org/viewtopic.php?pid=1684388

Last edited by Storm (2017-01-18 10:50:46)

oeut1341 · 2017-01-18 13:19:25

Storm wrote:

oeut1341 wrote:
Fresh core dump (~30 min ago): https://mega.nz/#!DdwjARCR!HctYG4Kwp7Kb … waeAmpJo78
For me that's an empty archive

Not if you decompress using bunzip2. (It's not a tar archive, but a single compressed file.)

Storm · 2017-01-18 15:16:17

In another thread someone suggested to roll back to the state before a specific Xorg update, and this seem to work, I should have crashed at least twice by now and I didn't.

Temp solution: https://bbs.archlinux.org/viewtopic.php … 4#p1683934

V1del · 2017-01-18 15:29:26

If you still try to go through with this further, have you tried simply removing your xorg.conf ? nvidia-xconfig generated plain xorg.conf have never been that useful and they might have completely broken something now. FWIW Xorg 1.19 has been completely stable here.

Storm · 2017-01-18 15:55:53

V1del wrote:

If you still try to go through with this further, have you tried simply removing your xorg.conf ? nvidia-xconfig generated plain xorg.conf have never been that useful and they might have completely broken something now. FWIW Xorg 1.19 has been completely stable here.

Yes I tried. Did not help.

Some people are affected, others are not and nobody seems to know why.

oeut1341 · 2017-01-18 19:09:15

Storm wrote:

In another thread someone suggested to roll back to the state before a specific Xorg update, and this seem to work, I should have crashed at least twice by now and I didn't.
Temp solution: https://bbs.archlinux.org/viewtopic.php … 4#p1683934

I rolled back xorg-server to 1.19.0-6 yesterday and haven't had a single crash ever since. So the problem is with 1.19.1-1, not 1.19 itself. Which is weird, considering it should include only fixes.

Last edited by oeut1341 (2017-01-18 19:18:39)

Storm · 2017-01-18 21:13:33

I just tested if the rollback fixes the SDDM crash too, and it does. This supports the theory that the two issues are related.

Storm · 2017-01-18 21:16:24

oeut1341 wrote:

I rolled back xorg-server to 1.19.0-6 yesterday and haven't had a single crash ever since. So the problem is with 1.19.1-1, not 1.19 itself. Which is weird, considering it should include only fixes.

Did you roll back only Xorg? As if yes, that proves it's Xorg itself.

Arch Linux

#1 2017-01-13 00:27:42

SDDM crash and random X crashes since upgrade

#2 2017-01-13 15:04:09

Re: SDDM crash and random X crashes since upgrade

#3 2017-01-13 20:01:21

Re: SDDM crash and random X crashes since upgrade

#4 2017-01-14 11:27:26

Re: SDDM crash and random X crashes since upgrade

#5 2017-01-14 12:01:35

Re: SDDM crash and random X crashes since upgrade

#6 2017-01-14 13:05:13

Re: SDDM crash and random X crashes since upgrade

#7 2017-01-14 13:32:11

Re: SDDM crash and random X crashes since upgrade

#8 2017-01-14 14:37:37

Re: SDDM crash and random X crashes since upgrade

#9 2017-01-14 16:09:05

Re: SDDM crash and random X crashes since upgrade

#10 2017-01-14 16:14:08

Re: SDDM crash and random X crashes since upgrade

#11 2017-01-14 18:44:10

Re: SDDM crash and random X crashes since upgrade

#12 2017-01-15 13:16:45

Re: SDDM crash and random X crashes since upgrade

#13 2017-01-15 13:24:01

Re: SDDM crash and random X crashes since upgrade

#14 2017-01-15 17:04:13

Re: SDDM crash and random X crashes since upgrade

#15 2017-01-16 15:22:20

Re: SDDM crash and random X crashes since upgrade

#16 2017-01-18 10:30:48

Re: SDDM crash and random X crashes since upgrade

#17 2017-01-18 10:38:05

Re: SDDM crash and random X crashes since upgrade

#18 2017-01-18 10:50:05

Re: SDDM crash and random X crashes since upgrade

#19 2017-01-18 13:19:25

Re: SDDM crash and random X crashes since upgrade

#20 2017-01-18 15:16:17

Re: SDDM crash and random X crashes since upgrade

#21 2017-01-18 15:29:26

Re: SDDM crash and random X crashes since upgrade

#22 2017-01-18 15:55:53

Re: SDDM crash and random X crashes since upgrade

#23 2017-01-18 19:09:15

Re: SDDM crash and random X crashes since upgrade

#24 2017-01-18 21:13:33

Re: SDDM crash and random X crashes since upgrade

#25 2017-01-18 21:16:24

Re: SDDM crash and random X crashes since upgrade

Board footer