You are not logged in.
Pages: 1
Dear Madams/Sirs
I would really appreciate help with this issue as it is one I am struggling to resolve:
-> After starting x lxde starts as normal.
-> pressing the logout button causes the system to hang and it can only be shutdown by holding the power button down (blind typing does not work as suggested elsewhere)
-> pressing the power button or typing reboot/shutdown in the terminal once in xorg also causes the system to hang rather than the desired effect.
-> The issue is independent of using nvidia-xconfig (seen that as a suggested solution elsewhere)
-> The issue is independent on if startx is run as root or not (as suggested in another thread)
-> If "exec startlxde" is omitted from the xinitrc file the system will hang upon starting x.
Please inform me if you require any more details and specify which logs/outputs you would like me to post. Thank you kindly for your assistance.
Offline
Welcome to the Arch Forums.
Is it a brand new install of Arch Linux? Laptop or desktop? Did you install lxde as root? Has it worked before?
Last edited by dockland (2016-08-03 06:20:47)
I possess a device, in my pocket, that is capable of accessing the entirety of information known to man.
I use it to look at funny pictures of cats and to argue with strangers.
Offline
1) is it a brand new install of Arch - Yes it is.
2) Laptop or Desktop - Desktop
3) Did you install lxde as root - I ran sudo pacman -S lxde to install it while logged on to an account in the wheel group.
4) Has it worked before - yes it has worked on my previous desktop (this is a new one) as well as my laptop and raspberry pi.
Offline
Are you in the power user group (if it's still needed in lxde, not sure. Have you re-installed upower just to check?
Last edited by dockland (2016-08-03 06:20:58)
I possess a device, in my pocket, that is capable of accessing the entirety of information known to man.
I use it to look at funny pictures of cats and to argue with strangers.
Offline
dockland: there is no need to quote the previous post every time you answer: it just bloats the thread and makes it much harder for people to read it.
You only need to quote if you wish to address specifics in the reply, and then only quote the necessary text, not the entire post.
Offline
The installation of upower had no effect. I don't think it is an issue with lxde, the power button should still work right? Also when I didn't include exec startlxde in the xinitrc the Xserver may be expected to throw an error perhaps? Instead that causes it to hang as well.
Offline
Have you tried using the acpi=force (or similar) kernel parameter?
Jin, Jîyan, Azadî
Offline
Also when I didn't include exec startlxde in the xinitrc the Xserver may be expected to throw an error perhaps? Instead that causes it to hang as well.
When the last command in xinintrc exits the XServer should terminate as well.
Could you supply the full journal output from a boot that hangs.
Incase it is xorg related please also supply if it exists /etc/X11/xorg.conf, any .conf file in /etc/X11/xorg.d/ the xorg.log file and the .xinitrc you are using.
Edit:
As you do not seem to have stated it explicitly, without starting the XServer does
$ systemctl poweroffor equivalent succeed?
Last edited by loqs (2016-08-02 21:16:18)
Offline
Shutdown works as normal when there is no xserver running.
here is my xinitrc
https://nopaste.me/view/cc228eff
here is xorg.0.log and xorg.1.log:
https://nopaste.me/view/b5beb607
https://nopaste.me/view/e160472a
I am aware it says "number of screens does not match number of devices configuration failed" but I thought that was a separate error? Also for the record I am using only 1 screen which is connected using HDMI to the nvidia graphics card.
There is no xorg.conf and no files in /etc/X11/xorg.conf.d however there are multiple files in /usr/share/X11/xorg.conf.d .
They are called as follows:
10-amdgpu.conf 10-quirks.conf 50-vmmouse.conf 70-synaptics.conf
10-evdev.conf 50-joystick.conf 60-libinput.conf nvidia-drm-outputclass.conf
would you be interested in seeing any of them?
Offline
with regards to acpi force, from what I understand from the wiki that may enable me to shutdown (it is used for power management right?) but will I be able to log off from X? Also is that a "hacky" solution that doesn't really address the underlying error?
(Rereading that it seems to come across a bit rude. I apologize if offence was caused, I am being inquisitive rather than condescending)
Offline
[ 15.153] (II) xfree86: Adding drm device (/dev/dri/card1)
[ 15.153] (II) systemd-logind: got fd for /dev/dri/card1 226:1 fd 8 paused 0
[ 15.153] (II) xfree86: Adding drm device (/dev/dri/card0)
[ 15.153] (II) systemd-logind: got fd for /dev/dri/card0 226:0 fd 9 paused 0
[ 15.154] (--) PCI: (0:0:2:0) 8086:0412:1458:d000 rev 6, Mem @ 0xf7400000/4194304, 0xd0000000/268435456, I/O @ 0x0000f000/64
[ 15.154] (--) PCI:*(0:1:0:0) 10de:13c2:19da:1366 rev 161, Mem @ 0xf6000000/16777216, 0xe0000000/268435456, 0xf0000000/33554432, I/O @ 0x0000e000/128, BIOS @ 0x????????/131072The system appears to have Hybrid Graphics ( speciffically Intel+Nvidia )
See also NVIDIA_Optimus Bumblebee and PRIME
Which one is of the above is system configured to use?
Could you supply the full journal output from a boot that hangs.
You omitted this item.
Systemd#Filtering_output
Edit:
Looking at the number of xorg video drivers listed in the xorg.0.log when you installed X did you install either the xorg group or the xorg-drivers group?
Last edited by loqs (2016-08-03 17:47:50)
Offline
I typed "sudo Xorg :0 --configure" to configure it so I am not sure sorry. Here is a jounalctl output for a session where I logged on started X and then pressed the logout button. Note to reboot I had to hold the power button.
https://nopaste.me/view/0d5909f8
Yes my computer is called ninja.
Offline
Aug 04 16:32:02 ninja systemd-logind[360]: System is powering down.
Aug 04 16:32:02 ninja systemd[1]: Stopped target Graphical Interface.As you are using startx would expect Graphical Interface ( graphical.target ) to be effectively unused and nothing is recorded about the stopping of multi-user.target or beyond.
Aug 04 16:31:25 ninja kernel: microcode: CPU0 sig=0x306c3, pf=0x2, revision=0x19
Aug 04 16:31:25 ninja kernel: microcode: CPU1 sig=0x306c3, pf=0x2, revision=0x19
Aug 04 16:31:25 ninja kernel: microcode: CPU2 sig=0x306c3, pf=0x2, revision=0x19
Aug 04 16:31:25 ninja kernel: microcode: CPU3 sig=0x306c3, pf=0x2, revision=0x19
Aug 04 16:31:25 ninja kernel: microcode: Microcode Update Driver: v2.01 <tigran@aivazian.fsnet.co.uk>, Peter OrubaThere is a microcode update available for that systems cpu.
$ bsdtar -Oxf /boot/intel-ucode.img | iucode_tool -tb -ls0x306c3 -
selected microcodes:
001: sig 0x000306c3, pf mask 0x32, 2016-03-16, rev 0x0020, size 22528Microcode#Enabling_Intel_microcode_updates
Would recommend you enable microcode updates, configure hybrid graphics as previously mentioned and see if that is enough to resolve the issue.
Offline
-> set up Nvidia optimus as per wiki page
-> enabled microcode updates (it appears to have updated as well)
-> problem still persists.
I greatly appreciate you taking the time to help me fix this. Is there any other files you want me to post? Bear in mind some of the previous ones have been updated.
Offline
Is there any other files you want me to post?
The xorg.conf plus xorg.log as they are now and journal of the latest stuck session please.
Keyboard_shortcuts#Kernel After enabling sysrq does using the unraw shortcut followed by the kill shortcut kill the X server and leave the system responsive?
If not does using the REISUB combination of shortcuts reboot the system ( please leave some time between them for the commands to work blindly )
Edit:
Other things to try install and use the linux-lts and nvidia-lts along side the already installed packages to see if it is a kernel issue.
More of a housekeeping issue
$ pacman -Qg xorg-driversYou can remove all the xf86-video-* results as the nvidia driver is provided by the nvidia package which is not a member of the group and the modesetting is provided by the xorg-server package also not a member of the group.
You can remove all the xf86-input-* results apart from xf86-input-libinput and or xf86-input-evdev. libinput is newer than evdev and will be used in preference if no settings are given in xorg.conf.
Edit2:
From the original xorg.log the display was connected to the Nvidia card. With the optimus setup did you move the connection to the motherboards output?
Last edited by loqs (2016-08-06 16:10:43)
Offline
I'm sorry to keep you waiting I will respond soon, I work two jobs and have just been so busy lately.
Offline
Ok here we go, sorry for the long wait, had problems with one of my partitions becoming corrupt as well as chrome refusing to work with youtube, but they are all sorted now.
-> Removed the stated files
-> nvidia-lts and linux-lts both installed
-> unraw followed by kill causes hanging
-> I did not move the connection to the motherboards output (or at least not deliberately)
journal:
https://nopaste.me/view/3b58464b
xorg.conf:
https://nopaste.me/view/69db329a
xorg.log:
https://nopaste.me/view/436fb793
Note I am not sure which session the log file is from (its just the xorg.0.log file), it's still hanging and I haven't changed anything recently so the system configuration should be the same for the journalctl and xorg.0.log files.
Offline
Depending on which bootloader you use you may have to manually create an entry for linux-lts then boot using that entry
You can check you have booted into the linux-lts kernel with
$ uname -rThe result should be similar to 4.4.20-1-lts-ARCH if it is the current lts kernel and 4.7.2-1-ARCH if is the current stable.
Would also try Head_on_a_Stick's suggestion of investigating the acpi kernel parameter to diagnose if the issue is related to acpi.
I missed that the system is a desktop with a discrete nvidia card in my earlier posts so as long as the display is connected to the discrete card you should be able to safely remove xorg.conf and install nvidia-libgl which will conflict/replace mesa-libgl which should give you hardware acceleration.
Would investigate in your firmware if there is an option to disable integrated graphics ( in case this is the cause of the issue )
Would also see if there is newer firmware available for your motherboard.
To rule out lxde as the cause would suggest uncommenting the twm, xclock and xterm lines in your .xinintc and comment out the startlxde line ( please ensure xorg-twm, xterm and xorg-xclock are installed first )
You can of course revert those changes once lxde has been rules out as the cause.
Offline
Definitely not LXDE as I tested what you suggested. uname -r suggests it is the current stable not the lts, is that an issue? Will try the other suggestions as well.
Offline
Pages: 1