latest update kills PLASMA

English Install/Boot/Login
1

hi,

with system as detailed below PLASMA is totally a none goer

laptop boots ok into tui and there runs ok

with PLASMA running the laptop response time is very slow with lots of disk activity.
inputs from keyboard or mouse only accepted perhaps every 100 seconds

laptop performance using iceWM is rock solid

any hints for a resolution welcomed

cheers

2019-09-06 11:02:59
System: Host: g7-2246sg Kernel: 5.2.11-1-default x86_64 bits: 64 Desktop: IceWM 1.5.4 Distro: openSUSE Tumbleweed 20190904
Machine: Type: Laptop System: Hewlett-Packard product: HP Pavilion g7 Notebook PC v: 088B110000305910000620100 serial: <filter>
Mobo: Hewlett-Packard model: 184B v: 57.50 serial: <filter> UEFI: Insyde v: F.17 date: 11/02/2012
CPU: Topology: Quad Core model: AMD A8-4500M APU with Radeon HD Graphics bits: 64 type: MCP L2 cache: 2048 KiB
Speed: 1345 MHz min/max: 1400/1900 MHz Core speeds (MHz): 1: 1194 2: 1369 3: 1372 4: 1381
Graphics: Device-1: Advanced Micro Devices [AMD/ATI] Trinity [Radeon HD 7640G] driver: radeon v: kernel
Display: x11 server: X.org 1.20.5 driver: radeon FAILED: ati unloaded: fbdev,modesetting,vesa resolution: <xdpyinfo missing>
OpenGL: renderer: AMD ARUBA (DRM 2.50.0 / 5.2.11-1-default LLVM 8.0.1) v: 4.3 Mesa 19.1.5

ps. Radeon driver installed and running ok
above failure indication is due to a failure in inxi.

2

Keep in mind,
YIf you’re installed on BTRFS, you’ll always have the option to roll back to just before you did the problem update.

Short of a rollback,
You have other options…

If you think your update was just broken… I’d recommend considering in the following order (You decide what you want to try, maybe not any or all of them)

  1. Try updating again. More or less assumes whatever caused your problem was transient so upgrading again might fix.
zypper dup
  1. Re-run your update but force packages to be replaced when a conflict might occur. See MAN pages, there is some risk involved but you seem to have a pretty convenient workound in the meantime by booting into IceWM fine.
zypper dup --replacefiles
  1. Install the older version of available files instead of the newest version. Like a rollback without actually doing a rollback
zypper dup --oldpackage

Good Luck,
TSU

3

I’m assuming that this is with the 20190904 update.

I’ll note that I am not having problems. So the best guess is that this is a video problem.

I’m not seeing anything in the 20190904 update that should have that effect. But maybe you skipped the 20190902 update, so have a new kernel.

The first thing to try, is booting from the previous kernel. That should be in the Advanced submenu of grub.

4

If the update previous to this one was a really long time ago it could be related this old issue possibly.

https://forums.opensuse.org/showthread.php/534273-Upgrade-to-20181218-killed-all-but-one-graphical-login

5

thanks for replying

ext4 is the formatting being used

If you think your update was just broken… I’d recommend considering in the following order (You decide what you want to try, maybe not any or all of them)

zypper dup was used for the original update and has been used multiple times since re-installing the updates

As iceWM is doing the job efficiently, I’ll wait for the next big change and see how things go.

thanks again

6

thanks for replying

after the update neither xf86-video-intel or xf86-video-nouveau are installed

the installed drivers are,
xf86-video-amdgpu-19.0.1-2.1.x86_64
xf86-video-ati-19.0.1-1.4.x86_64
xf86-video-fbdev-0.5.0-1.7.x86_64
xf86-video-mach64-6.9.6-1.7.x86_64
xf86-video-r128-6.11.0-2.2.x86_64
xf86-video-vesa-2.4.0-1.7.x86_64

thanks again

7

just for info,
the following three lines show first on the display on each boot,

Sep 07 04:26:09 do_IRQ: 1.55 No irq handler for this vector
Sep 07 04:26:09 do_IRQ: 2.55 No irq handler for this vector
Sep 07 04:26:09 do_IRQ: 3.55 No irq handler for this vector

not sure how to interpret these statements

8

https://duckduckgo.com/?q=No+irq+handler+for+this+vector

9

Any clues in systemd journalctl for issues with Plasma?

10

You can filter:

erlangen:~ # journalctl -b _PID=1306 --grep error
-- Logs begin at Tue 2019-09-03 21:44:16 CEST, end at Sat 2019-09-07 16:39:59 CEST. --
Sep 07 13:31:10 erlangen plasmashell[1306]: libkcups: CUPS-Get-Printers last error: 0 successful-ok
Sep 07 13:31:10 erlangen plasmashell[1306]: libkcups: Get-Jobs last error: 0 successful-ok
Sep 07 13:31:10 erlangen plasmashell[1306]: libkcups: Get-Jobs last error: 0 successful-ok
Sep 07 13:31:10 erlangen plasmashell[1306]: libkcups: Create-Printer-Subscriptions last error: 0 successful-ok
Sep 07 13:43:23 erlangen plasmashell[1306]: file:///usr/share/plasma/shells/org.kde.plasma.desktop/contents/applet/CompactApplet.qml:64: TypeError: Cannot read property 'Layout' of null
Sep 07 13:43:23 erlangen plasmashell[1306]: file:///usr/share/plasma/shells/org.kde.plasma.desktop/contents/applet/CompactApplet.qml:82: TypeError: Type error
Sep 07 13:43:23 erlangen plasmashell[1306]: libkcups: Cancel-Subscription last error: 0 successful-ok
Sep 07 13:43:23 erlangen plasmashell[1306]: libkcups: Create-Printer-Subscriptions last error: 0 successful-ok
Sep 07 13:43:23 erlangen plasmashell[1306]: libkcups: Cancel-Subscription last error: 0 successful-ok
erlangen:~ #
11

having pci=nomsi,noaer on the kernel cmd line did not have any effect

it was tried with secure boot/ support for secure boot both enabled and disabled

thanks for the heads up

12

for one boot, with cmd journalctl -g _PID=1306 --grep error, the output is,

-- Reboot --
Sep 08 01:52:12 g7-2246sg ksmserver[2103]: Qt: Session management error: networkIdsList argument is NULL
Sep 08 01:52:13 g7-2246sg kscreen_backend_launcher[2109]: kscreen.xcb.helper: Event Error:  147
Sep 08 01:52:16 g7-2246sg kwin_x11[2107]: qt.qpa.xcb: QXcbConnection: XCB error: 3 (BadWindow), sequence: 828, resource id: 14680069, major code: 18 (ChangeProperty), minor code: 0
Sep 08 01:52:31 g7-2246sg plasmashell[2118]: libkcups: CUPS-Get-Printers last error: 1030 No destinations added.
Sep 08 01:52:31 g7-2246sg plasmashell[2118]: libkcups: Create-Printer-Subscriptions last error: 0 successful-ok
Sep 08 01:52:32 g7-2246sg plasmashell[2118]: libkcups: Get-Jobs last error: 0 successful-ok
Sep 08 01:52:32 g7-2246sg plasmashell[2118]: libkcups: Get-Jobs last error: 0 successful-ok
Sep 08 01:52:32 g7-2246sg kwin_x11[2107]: qt.qpa.xcb: QXcbConnection: XCB error: 3 (BadWindow), sequence: 1094, resource id: 6291492, major code: 18 (ChangeProperty), minor code: 0
Sep 08 01:52:43 g7-2246sg pulseaudio[2188]: E: [pulseaudio] bluez5-util.c: GetManagedObjects() failed: org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.
Sep 08 01:53:02 g7-2246sg plasmashell[2118]: file:///usr/share/plasma/plasmoids/org.kde.plasma.taskmanager/contents/ui/Task.qml:375: TypeError: Cannot read property 'length' of undefined
Sep 08 01:53:03 g7-2246sg plasmashell[2118]: file:///usr/share/plasma/plasmoids/org.kde.plasma.taskmanager/contents/ui/main.qml:454: TypeError: Type error
Sep 08 01:53:05 g7-2246sg plasmashell[2118]: file:///usr/share/plasma/plasmoids/org.kde.plasma.taskmanager/contents/ui/main.qml:454: TypeError: Type error
Sep 08 01:53:06 g7-2246sg plasmashell[2118]: file:///usr/share/plasma/plasmoids/org.kde.plasma.taskmanager/contents/ui/Task.qml:375: TypeError: Type error
Sep 08 01:53:08 g7-2246sg kdeinit5[2301]: Qt: Session management error: networkIdsList argument is NULL
-- Reboot --

cheers

13
  • Is there a problem with kdeinit?
karl@erlangen:~/Downloads> journalctl -b --user --grep kdeinit5
-- Logs begin at Tue 2019-09-03 21:44:16 CEST, end at Sun 2019-09-08 09:09:57 CEST. --
Sep 07 13:31:08 erlangen dbus-daemon[1200]: [session uid=1000 pid=1200] Activating service name='org.kde.kglobalaccel' requested by ':1.6' (uid=1000 pid=1264 comm="kded5 [kdeinit5]                                  ")
Sep 07 13:31:08 erlangen dbus-daemon[1200]: [session uid=1000 pid=1200] Activating service name='org.kde.KScreen' requested by ':1.6' (uid=1000 pid=1264 comm="kded5 [kdeinit5]                                  ")
Sep 07 13:31:09 erlangen dbus-daemon[1200]: [session uid=1000 pid=1200] Activating service name='org.bluez.obex' requested by ':1.6' (uid=1000 pid=1264 comm="kded5 [kdeinit5]                                  ")
Sep 07 13:43:51 erlangen dbus-daemon[1200]: [session uid=1000 pid=1200] Activating service name='org.kde.kglobalaccel' requested by ':1.153' (uid=1000 pid=4322 comm="kded5 [kdeinit5]                                  ")
Sep 07 13:43:51 erlangen dbus-daemon[1200]: [session uid=1000 pid=1200] Activating service name='org.kde.KScreen' requested by ':1.153' (uid=1000 pid=4322 comm="kded5 [kdeinit5]                                  ")
karl@erlangen:~/Downloads>
  • What about with a pristine login (new user with empty home directory)?
14

hi,

the output of cmd, journalctl -g _PID=1306 --grep kde | grep fail, is

Sep 05 23:17:31 g7-2246sg kglobalaccel5[3982]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:31 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:43 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:43 g7-2246sg kglobalaccel5[4050]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:43 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:43 g7-2246sg kglobalaccel5[4055]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:45 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:45 g7-2246sg kglobalaccel5[4060]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:46 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:46 g7-2246sg kglobalaccel5[4066]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:46 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6
Sep 05 23:17:46 g7-2246sg kglobalaccel5[4072]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:47 g7-2246sg kglobalaccel5[4077]: This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Sep 05 23:17:47 g7-2246sg dbus-daemon[1766]: [session uid=1001 pid=1766] Activated service ‘org.kde.kglobalaccel’ failed: Process org.kde.kglobalaccel received signal 6



re-installing, kglobalaccel5 and dbus-1, have made no difference

this is a very new net install, but a new user will be tried when time allows

if cmd fsck is run after a PLASMA session lots of fixes are needed, no fixes are required after IceWM session

cheers
15

Run smartctl to check disk You may have localized disk problem in the area wher KDE is stored.

16

You need to fix /home. Watch for damage: smartctl -a /dev/sda, smartctl -t long /dev/sda.

17

hi,

both commands run, the second one took 244 minutes to complete,
no errors messages were reported in the journal or the display.

18

To make sure run fsck and smartctl for /. You need to boot into an alternate OS for fsck.

19

hi,

with a new user account all is fine

thanks again to all who responded

cheers