[opensuse-support] WTF, why does systemd-logind power down my computer?
Just happened the second time. I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it! The log reveals Aug 23 12:35:51 woodstock.pitnet kdeinit5[4101]: kdeinit5: sending SIGTERM to children. Aug 23 12:35:51 woodstock.pitnet kdeinit5[4101]: kdeinit5: Exit. Aug 23 12:35:51 woodstock.pitnet kwin_x11[4139]: The X11 connection broke (error 1). Did the X11 server die? Aug 23 12:35:51 woodstock.pitnet kscreen_backend_launcher[4138]: The X11 connection broke (error 1). Did the X11 server die? Aug 23 12:35:51 woodstock.pitnet systemd-logind[960]: System is powering down. Is that the latest "invention" of the systemd crew? -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2019-08-23 at 13:10 +0100, Peter Suetterlin wrote:
Just happened the second time.
I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it!
It should neither restart nor shut down the system. Just kill the X server. I don't dare try to replicate, but I'll try keep it in mind to try after doing the updates.
The log reveals
Aug 23 12:35:51 woodstock.pitnet kdeinit5[4101]: kdeinit5: sending SIGTERM to children. Aug 23 12:35:51 woodstock.pitnet kdeinit5[4101]: kdeinit5: Exit. Aug 23 12:35:51 woodstock.pitnet kwin_x11[4139]: The X11 connection broke (error 1). Did the X11 server die? Aug 23 12:35:51 woodstock.pitnet kscreen_backend_launcher[4138]: The X11 connection broke (error 1). Did the X11 server die? Aug 23 12:35:51 woodstock.pitnet systemd-logind[960]: System is powering down.
Is that the latest "invention" of the systemd crew?
Report in bugzilla, I would say. - -- Cheers, Carlos E. R. (from openSUSE 15.0 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCXV/abhwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVT1AAn2avkqYZqrZdUbSFdCz4 L6+UcA4pAKCHP+uvx7vodEyLv14iCRvI8WdO9w== =Idms -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Carlos E. R. wrote:
On Friday, 2019-08-23 at 13:10 +0100, Peter Suetterlin wrote:
I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it!
It should neither restart nor shut down the system. Just kill the X server.
Sloppy wording. The expected thing would be the login manager restarts the X server and fires up the login screen, which I abbreviated as 'restarting it'.
Report in bugzilla, I would say.
I have no reliable way to trigger it, sometimes it *does* as I expect. Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....). -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Am Freitag, 23. August 2019, 14:57:20 CEST schrieb Peter Suetterlin:
Carlos E. R. wrote:
On Friday, 2019-08-23 at 13:10 +0100, Peter Suetterlin wrote:
I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it!
It should neither restart nor shut down the system. Just kill the X server.
Sloppy wording. The expected thing would be the login manager restarts the X server and fires up the login screen, which I abbreviated as 'restarting it'.
Report in bugzilla, I would say.
I have no reliable way to trigger it, sometimes it *does* as I expect. Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....).
Well, Peter, while I understand your sediments, nobody can reply something adequately without knowing the basic facts, that would be openSUSE version, X driver and window manager in use. I never experienced this, and I'm using this once or twice a month on different machines... Well, my son twice a week, since he's hitting his machine hard with blender 2.80 (oh, rendering 100.000.000 vertices was may be a bit too much for 8 GB DRAM...). Cheers, Pete -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Hmm, somehow our mail server ended up on spamhaus, and the reply somewhere else :o Trying again Hans-Peter Jansen wrote:
Well, Peter, while I understand your sediments, nobody can reply something adequately without knowing the basic facts, that would be openSUSE version, X driver and window manager in use.
Ooops. Indeed, I didn't specify anything. My excuses. I must have benn *really* embarassed.... This is Tumbleweed, latest version (as of Friday, i.e. 20190820), KDE/Plasma with sddm. I had tried two more times, it powered down both times, so there's a bugreport now: https://bugzilla.opensuse.org/show_bug.cgi?id=1147111
I never experienced this, and I'm using this once or twice a month on different machines... Well, my son twice a week, since he's hitting his machine hard with blender 2.80 (oh, rendering 100.000.000 vertices was may be a bit too much for 8 GB DRAM...).
Yes, and it's as 'hardcoded' in my fingers as CTRL-ALT-DEL and similar stuff. Nothing one should change... -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Am Montag, 26. August 2019, 20:54:47 CEST schrieb Peter Suetterlin:
Hmm, somehow our mail server ended up on spamhaus, and the reply somewhere else :o Trying again
Cool, I had such issues with ionos 1&1 mail servers being blacklisted, when writing to governmental organizations in the baltic states... That's pure fun.
Hans-Peter Jansen wrote:
Well, Peter, while I understand your sediments, nobody can reply something adequately without knowing the basic facts, that would be openSUSE version, X driver and window manager in use.
Ooops. Indeed, I didn't specify anything.
My excuses. I must have benn *really* embarassed....
Hehe. I feel with you.
This is Tumbleweed, latest version (as of Friday, i.e. 20190820), KDE/Plasma with sddm.
What GFX card?
I had tried two more times, it powered down both times, so there's a bugreport now: https://bugzilla.opensuse.org/show_bug.cgi?id=1147111
Okay, want to show us the output of: rpm -V $(rpm -qa | grep systemd) and rpm -V $(rpm -qa | grep x11) Cheers, Pete -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Hans-Peter Jansen wrote:
This is Tumbleweed, latest version (as of Friday, i.e. 20190820), KDE/Plasma with sddm.
What GFX card?
For normal desktop, an Skylake integrated HD530. The laptop has an Optimus (940MX), but I only rarely use it (with bumblebee/optirun)
I had tried two more times, it powered down both times, so there's a bugreport now: https://bugzilla.opensuse.org/show_bug.cgi?id=1147111
Okay, want to show us the output of:
rpm -V $(rpm -qa | grep systemd)
.M....... c /etc/X11/xorg.conf.d/00-keyboard.conf .M....... c /etc/locale.conf .M....... c /etc/machine-id S.5....T. c /etc/systemd/journald.conf S.5....T. c /etc/systemd/system.conf S.5....T. c /etc/systemd/timesyncd.conf .M....... c /etc/vconsole.conf .M....... g /var/lib/systemd/random-seed S.5....T. c /etc/systemd/coredump.conf The only change in system.conf is < DefaultTimeoutStartSec=30s < DefaultTimeoutStopSec=10s which is already active since more than a year.
and
rpm -V $(rpm -qa | grep x11)
Only font stuff FWIW: woodstock:~% setxkbmap -query rules: evdev model: microsoftpro layout: us options: terminate:ctrl_alt_bksp -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Am Freitag, 30. August 2019, 10:52:36 CEST schrieb Peter Suetterlin:
Hans-Peter Jansen wrote:
This is Tumbleweed, latest version (as of Friday, i.e. 20190820), KDE/Plasma with sddm.
What GFX card?
For normal desktop, an Skylake integrated HD530. The laptop has an Optimus (940MX), but I only rarely use it (with bumblebee/optirun)
Maybe, I missed something, but do you suffer from this on multiple machines?
I had tried two more times, it powered down both times, so there's a bugreport now: https://bugzilla.opensuse.org/show_bug.cgi?id=1147111
Okay, want to show us the output of:
rpm -V $(rpm -qa | grep systemd)
.M....... c /etc/X11/xorg.conf.d/00-keyboard.conf .M....... c /etc/locale.conf .M....... c /etc/machine-id S.5....T. c /etc/systemd/journald.conf S.5....T. c /etc/systemd/system.conf S.5....T. c /etc/systemd/timesyncd.conf .M....... c /etc/vconsole.conf .M....... g /var/lib/systemd/random-seed S.5....T. c /etc/systemd/coredump.conf
The only change in system.conf is < DefaultTimeoutStartSec=30s < DefaultTimeoutStopSec=10s which is already active since more than a year.
and
rpm -V $(rpm -qa | grep x11)
Only font stuff
FWIW: woodstock:~% setxkbmap -query rules: evdev model: microsoftpro layout: us options: terminate:ctrl_alt_bksp
Have you tried to use another keyboard model? pc101 perhaps? It might boil down to a key mapping issue. I vaguely remember some backspace vs. delete mapping issues in ancient times... Cheers, Pete -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
Hans-Peter Jansen wrote:
For normal desktop, an Skylake integrated HD530. The laptop has an Optimus (940MX), but I only rarely use it (with bumblebee/optirun)
Maybe, I missed something, but do you suffer from this on multiple machines?
I have one other machine, that seems to work properly. Two other ones are headless...
FWIW: woodstock:~% setxkbmap -query rules: evdev model: microsoftpro layout: us options: terminate:ctrl_alt_bksp
Have you tried to use another keyboard model? pc101 perhaps? It might boil down to a key mapping issue. I vaguely remember some backspace vs. delete mapping issues in ancient times...
Guess those were automatic values from setup. I haven't changed things there, and would have to start looking where to change this. And it was working fine for 3 years..... -- To unsubscribe, e-mail: opensuse-support+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-support+owner@opensuse.org
On 23/08/2019 14.57, Peter Suetterlin wrote:
Carlos E. R. wrote:
On Friday, 2019-08-23 at 13:10 +0100, Peter Suetterlin wrote:
I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it!
It should neither restart nor shut down the system. Just kill the X server.
Sloppy wording. The expected thing would be the login manager restarts the X server and fires up the login screen, which I abbreviated as 'restarting it'.
Ok :-)
Report in bugzilla, I would say.
I have no reliable way to trigger it, sometimes it *does* as I expect.
Ah.
Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....).
I'll try on next occasion :-) I do not want to try on a virtual machine, because sometimes it is the host which responds :-( -- Cheers / Saludos, Carlos E. R. (from 15.0 x86_64 at Telcontar)
W dniu 23.08.2019 o 16:13, Carlos E. R. pisze:
On 23/08/2019 14.57, Peter Suetterlin wrote:
Carlos E. R. wrote:
On Friday, 2019-08-23 at 13:10 +0100, Peter Suetterlin wrote:
I had pressed CTRL-ALT-Backspace twice to kill the X server - and the system shuts down instead of restarting it!
It should neither restart nor shut down the system. Just kill the X server.
Sloppy wording. The expected thing would be the login manager restarts the X server and fires up the login screen, which I abbreviated as 'restarting it'.
Ok :-)
Report in bugzilla, I would say.
I have no reliable way to trigger it, sometimes it *does* as I expect.
Ah.
Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....).
I'll try on next occasion :-)
I do not want to try on a virtual machine, because sometimes it is the host which responds :-(
Virtual machines usually have some ways of sending "dangerous" key combinations. VirtualBox: "Host key"+Backspace (https://www.virtualbox.org/manual/UserManual.html#host-key-customize) Virt-manager: you can see on the screenshot, that there's "Send key" menu (https://www.virt-manager.org/wp-content/uploads/2014/01/console.png)
On 23/08/2019 18.38, Adam Mizerski wrote:
W dniu 23.08.2019 o 16:13, Carlos E. R. pisze:
On 23/08/2019 14.57, Peter Suetterlin wrote:
Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....).
I'll try on next occasion :-)
I do not want to try on a virtual machine, because sometimes it is the host which responds :-(
Virtual machines usually have some ways of sending "dangerous" key combinations.
VirtualBox: "Host key"+Backspace (https://www.virtualbox.org/manual/UserManual.html#host-key-customize)
TypeCABS Backspace Inject Ctrl+Alt+Backspace does nothing, except deleting chars in the guest running apps, as expected from backspace. There is an entry in the menu which says it is host+backspace. "host" mean the host key combination. I tried that, and the HOST desktop died, not the guest. As the host desktop died, it also killed the virtualmachine as if pulling the cable. This is disastrous. -- Cheers / Saludos, Carlos E. R. (from 15.0 x86_64 at Telcontar)
W dniu 23.08.2019 o 20:13, Carlos E. R. pisze:
On 23/08/2019 18.38, Adam Mizerski wrote:
W dniu 23.08.2019 o 16:13, Carlos E. R. pisze:
On 23/08/2019 14.57, Peter Suetterlin wrote:
Therefore I tried asking here if others have seen it (assuming that more people are reading forum posts than bugreports....).
I'll try on next occasion :-)
I do not want to try on a virtual machine, because sometimes it is the host which responds :-(
Virtual machines usually have some ways of sending "dangerous" key combinations.
VirtualBox: "Host key"+Backspace (https://www.virtualbox.org/manual/UserManual.html#host-key-customize)
TypeCABS Backspace Inject Ctrl+Alt+Backspace
does nothing, except deleting chars in the guest running apps, as expected from backspace.
There is an entry in the menu which says it is host+backspace. "host" mean the host key combination. I tried that, and the HOST desktop died, not the guest. As the host desktop died, it also killed the virtualmachine as if pulling the cable.
This is disastrous.
"Host key" is displayed next to status icons and by default it's right control. https://www.virtualbox.org/manual/UserManual.html#keyb_mouse_normal
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2019-08-23 at 20:20 +0200, Adam Mizerski wrote:
W dniu 23.08.2019 o 20:13, Carlos E. R. pisze:
On 23/08/2019 18.38, Adam Mizerski wrote:
W dniu 23.08.2019 o 16:13, Carlos E. R. pisze:
On 23/08/2019 14.57, Peter Suetterlin wrote:
Virtual machines usually have some ways of sending "dangerous" key combinations.
VirtualBox: "Host key"+Backspace (https://www.virtualbox.org/manual/UserManual.html#host-key-customize)
TypeCABS Backspace Inject Ctrl+Alt+Backspace
does nothing, except deleting chars in the guest running apps, as expected from backspace.
There is an entry in the menu which says it is host+backspace. "host" mean the host key combination. I tried that, and the HOST desktop died, not the guest. As the host desktop died, it also killed the virtualmachine as if pulling the cable.
This is disastrous.
"Host key" is displayed next to status icons and by default it's right control. https://www.virtualbox.org/manual/UserManual.html#keyb_mouse_normal
I know. I changed it to be ctrl-alt, same as in vmware. The combo was seen by the host desktop and it acted on it, not by the VM environment. - -- Cheers, Carlos E. R. (from openSUSE 15.0 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHIEARECADIWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCXWBDfhQccm9iaW4ubGlz dGFzQGdteC5lcwAKCRC1MxgcbY1H1WJYAJ49r4QqabaCxf1hPN0HC2JGdUcchACe KEZ8SmPvTA2o6rZXLGfMD97rVEc= =/zE2 -----END PGP SIGNATURE-----
participants (5)
-
Adam Mizerski -
Carlos E. R. -
Carlos E. R. -
Hans-Peter Jansen -
Peter Suetterlin