PDA

View Full Version : [kubuntu] Blackscreen during startup



kuv02
January 23rd, 2016, 10:59 PM
When I reboot my uncles laptop, from time to time (like 20-30% of the time) it will hang in boot on a black screen.

I still can enter into Ctrl-ALT-F1 but executing startx from there will result in another blackscreen (this time with disabled backlight). I could switch to the next console, but I'm unable to start the grafic UI.

After such an event I started the OS sucessfully and imediatly checked the system log for errors. This is what I found when searching for "error" in KSystemLog:

23.01.2016 22:47:51 rf-acer kernel [ 4.747169] EXT4-fs (dm-0): re-mounted. Opts: errors=remount-ro
23.01.2016 22:47:51 rf-acer whoopsie[825] [22:47:51] GDBus.Error:org.freedesktop.DBus.Error.ServiceUnkn own: The name org.freedesktop.NetworkManager was not provided by any .service files
23.01.2016 22:47:53 rf-acer NetworkManager[834] <error> [1453585673.412844] [dns-manager/nm-dns-dnsmasq.c:398] update(): dnsmasq owner not found on bus: Could not get owner of name 'org.freedesktop.NetworkManager.dnsmasq': no such name
23.01.2016 22:47:57 rf-acer org.kde.kded5[1189] Qt: Session management error: networkIdsList argument is NULL
23.01.2016 22:47:57 rf-acer org.kde.kded5[1189] Qt: Session management error: networkIdsList argument is NULL
23.01.2016 22:48:07 rf-acer kernel [ 21.176439] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.460476] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.488526] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.512572] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.604686] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:09 rf-acer kernel [ 22.650318] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:09 rf-acer kernel [ 22.674300] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!

Do you think one of those messages could be the reason for the blackscreen? And what can I do about it?

matt_symes
January 23rd, 2016, 11:10 PM
Hi



After such an event I started the OS sucessfully and imediatly checked the system log for errors. This is what I found when searching for "error" in KSystemLog:

23.01.2016 22:47:51 rf-acer kernel [ 4.747169] EXT4-fs (dm-0): re-mounted. Opts: errors=remount-ro
23.01.2016 22:47:51 rf-acer whoopsie[825] [22:47:51] GDBus.Error:org.freedesktop.DBus.Error.ServiceUnkn own: The name org.freedesktop.NetworkManager was not provided by any .service files
23.01.2016 22:47:53 rf-acer NetworkManager[834] <error> [1453585673.412844] [dns-manager/nm-dns-dnsmasq.c:398] update(): dnsmasq owner not found on bus: Could not get owner of name 'org.freedesktop.NetworkManager.dnsmasq': no such name
23.01.2016 22:47:57 rf-acer org.kde.kded5[1189] Qt: Session management error: networkIdsList argument is NULL
23.01.2016 22:47:57 rf-acer org.kde.kded5[1189] Qt: Session management error: networkIdsList argument is NULL
23.01.2016 22:48:07 rf-acer kernel [ 21.176439] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.460476] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.488526] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.512572] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:07 rf-acer kernel [ 21.604686] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:09 rf-acer kernel [ 22.650318] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!
23.01.2016 22:48:09 rf-acer kernel [ 22.674300] [drm:gen8_irq_handler [i915_bpo]] *ERROR* The master control interrupt lied (SDE)!

Do you think one of those messages could be the reason for the blackscreen? And what can I do about it?

I would say it's almost definitely due to the last errors i highlighted above.

This may be a known bug and may be fixed by a kernel update.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1520040

Can you post the output of


cat /etc/lsb-release

What kernel are you booted into ?


uname -a


sudo lshw -C display

Kind regards

kuv02
January 23rd, 2016, 11:14 PM
Here it is:
~$ cat /etc/lsb-release
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=15.04
DISTRIB_CODENAME=vivid
DISTRIB_DESCRIPTION="Ubuntu 15.04"
~$ uname -a
Linux rf-acer 3.19.0-47-generic #53-Ubuntu SMP Mon Jan 18 14:02:48 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
~$ sudo lshw -C display
[sudo] password for user:
*-display
Beschreibung: VGA compatible controller
Produkt: Broadwell-U Integrated Graphics
Hersteller: Intel Corporation
Physische ID: 2
Bus-Informationen: pci@0000:00:02.0
Version: 09
Breite: 64 bits
Takt: 33MHz
Fähigkeiten: msi pm vga_controller bus_master cap_list rom
Konfiguration: driver=i915_bpo latency=0
Ressourcen: irq:53 memory:c2000000-c2ffffff memory:d0000000-dfffffff ioport:5000(Größe=64)
*-display
Beschreibung: 3D controller
Produkt: GM108M [GeForce 840M]
Hersteller: NVIDIA Corporation
Physische ID: 0
Bus-Informationen: pci@0000:04:00.0
Version: a2
Breite: 64 bits
Takt: 33MHz
Fähigkeiten: pm msi pciexpress bus_master cap_list
Konfiguration: driver=nvidia latency=0
Ressourcen: irq:55 memory:c3000000-c3ffffff memory:b0000000-bfffffff memory:c0000000-c1ffffff ioport:3000(Größe=128)

matt_symes
January 23rd, 2016, 11:37 PM
Hi


Produkt: Broadwell-U Integrated Graphics

There looks to be a bug in the broadwell chip and some of the kernels.

My advice would be to read the bug report i linked to above and this one.

https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1488719

Then you need to either upgrade or downgrade to a different kernel version until you find one that works as the 3.19 kernels seem to be a problem.

http://kernel.ubuntu.com/~kernel-ppa/mainline/

Maybe try this version first

http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.4-wily/

If you need instruction on how to update your kernel then please post back.

Do you have an external monitor connected ?

EDIT:

Apparently the 3.13 series of kernel may not suffer from this bug as you may want to roll back to that one.

Kind regards

kuv02
January 24th, 2016, 12:32 AM
There was an external monitor (TV) connected. I tried the boot with TV connected, because I thought, maybe the output is redirtected to the HDMI port.

Maybe I can get my uncle (owner of that laptop) to hold out until 16.04 comes out (it will have Kernel 4.4)

But I've read in one bug that the bug prevailed, in another I've read it was fixed. So not sure, if Kernel 4.4 will fix the problem.

How hard is it to switch back to an older kernel? And would this mean security problems?

matt_symes
January 24th, 2016, 09:28 PM
Hi


How hard is it to switch back to an older kernel?

No just reinstall the the linux genric kernel metapackge and the latest kernel from that kernel series.

You'll lose the benefits of newer kernels such as better support for new hardware though.


and would this mean security problems?

Not if you still use a supported kernel series and you keep that kernel updated. You are currently using the Vivid hardware enablement stack.

I can really only suggest you try kernels and see which ones work and update to 16.04 when you can.

Kind regards

kuv02
January 27th, 2016, 07:28 AM
Thanks
When I install an older Kernel, can I make it the default in grub, or will it only appear as one of the "alternative boot options"?

matt_symes
January 30th, 2016, 10:01 PM
Hi


Thanks
When I install an older Kernel, can I make it the default in grub, or will it only appear as one of the "alternative boot options"?

Sorry for the late reply.

When you install an older kernel, it should update grub and appear as the first (default) entry in your gub menu.

You can double check by booting into the grub's default entry, opening a terminal and typing


uname -r

I'm not sure who long the 1.13 series of kernels are going to be supported for though. It may be for the lifetime of Trusty, i'm not sure.

Kind regards