I have two systems that were recently upgraded from Fedora 26 to 27 using DNF system-upgrade. Both were successfully running a KDE desktop for several generations previously, including on F26. In both cases, after the upgrade, starting KDE showed the splash screen and then a blank screen with only a pointer cursor. The keyboard and mouse worked but the screen remained blank. Both systems are set to start in systemd's multi-user mode; KDE is started with the startx command.
One of the systems is a 32-bit laptop with little local content. I chose to simply do a fresh install on this one to get it working quickly. Fedora doesn't offer an installation image for 32-bit F27 so I installed F25 and then did successive upgrades to F26 and then F27. KDE worked in both F25 and F26 but had the same problem once I got to F27. I did it again and stopped at F26 to have a working system.
The other system is a modern 64-bit desktop with 8GB RAM. This one still has the problem. I'd like to find the cause and a solution before I make the effort to revert to F26 because this one has a huge amount of local content which I would have to backup and restore. (I do backups but more for recovery of individual files; a general recovery could have other problems.)
Note that Gnome works on this machine (but I hate Gnome).
The logs haven't given me any clue as to the cause of the problem. I have now run "startx &>logfile" hoping that might tell me something. After the blank screen appeared, I killed xinit and Xorg from a different console. The logfile is included below in the hope it might spark some insight from the group.
xauth: file /home/dave/.serverauth.1517 does not exist X.Org X Server 1.19.6 Release Date: 2017-12-20 X Protocol Version 11, Revision 0 Build Operating System: 4.14.11-300.fc27.x86_64 Current Operating System: Linux dc7800.compata.com 4.14.16-300.fc27.x86_64 #1 SMP Wed Jan 31 19:24:27 UTC 2018 x86_64 Kernel command line: BOOT_IMAGE=/vmlinuz-4.14.16-300.fc27.x86_64 root=/dev/mapper/fedora_dc7800-root ro rd.lvm.lv=fedora_dc7800/root rd.lvm.lv=fedora_dc7800/swap vconsole.font=latarcyrheb-sun16 biosdevname=0 net.ifnames=0 selinux=0 LANG=en_US.UTF-8 Build Date: 10 January 2018 11:00:12PM Build ID: xorg-x11-server 1.19.6-2.fc27 Current version of pixman: 0.34.0 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/home/dave/.local/share/xorg/Xorg.0.log", Time: Tue Feb 6 21:11:34 2018 (==) Using config file: "/etc/X11/xorg.conf" (==) Using config directory: "/etc/X11/xorg.conf.d" (==) Using system config directory "/usr/share/X11/xorg.conf.d" startkde: Starting up... dbus-update-activation-environment: warning: error sending to systemd: org.freedesktop.DBus.Error.Spawn.ChildExited: Process org.freedesktop.systemd1 exited with status 1 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/akregator.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/blogilo.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/disable_kmix.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_akregator.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_global.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_global_toolbar.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_kate.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/importwizard.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kaddressbook.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kalarm.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/katomic-levelset.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kcalcrc.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kget.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kgthemeprovider-migration.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kleopatra.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/klickety.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kmail.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kmail2.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/knavalbattle.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/knotes.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kontact.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/korganizer.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/krdb_libpathwipe.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/krunnerplugins.upd' for new updates 2018-02-06T21:12:02 krunnerplugins.upd: Found new update '5.9KRunnerPlugins' 2018-02-06T21:12:02 krunnerplugins.upd:3:'Script=krunnerplugins': Script 'krunnerplugins' not found 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/kscreenlocker.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/messageviewer.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/okular.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/pimsettingexporter.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/sieveeditor.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/webengineurlinterceptoradblock.upd' for new updates vmware-user: could not open /proc/fs/vmblock/dev Service started, version: 7.0.0 Service started, version: 7.0.0 Service started, version: 7.0.0 OpenGL vendor string: Intel Open Source Technology Center OpenGL renderer string: Mesa DRI Intel(R) Q35 OpenGL version string: 1.4 Mesa 17.2.4 OpenGL shading language version string: 1.20 Driver: Intel GPU class: i915/i945 OpenGL version: 1.4 GLSL version: 1.20 Mesa version: 17.2.4 Linux kernel version: 4.14.16 Requires strict binding: yes GLSL shaders: limited Texture NPOT support: yes Virtual Machine: no (II) AIGLX: Suspending AIGLX clients for VT switch (krb5-auth-dialog:2069): libnotify-WARNING **: Failed to connect to proxy (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Failed to read server caps (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Unsupported cache type for 'KEYRING:persistent:1000' (krb5-auth-dialog:2069): KrbAuthDialog-CRITICAL **: monitor_ccache: assertion 'ccache_name != NULL' failed Killed XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0" after 27 requests (27 known processed) with 0 events remaining. klauncher: Exiting on signal 1 Unexpected response from KInit (response = 0). startkde: Could not start ksmserver. Check your installation. -- I think this is where KDE hung and the following is after I killed it. Error: Can't open display: :0 KCrash: Application 'ksmserver' crashing... KCrash: Attempting to start /usr/libexec/drkonqi from kdeinit Warning: connect() failed: : No such file or directory KCrash: Attempting to start /usr/libexec/drkonqi directly Could not connect to D-Bus server: org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-SJIFqD8xwP: Connection refused startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
On 07/02/18 10:57 AM, Dave Close wrote:
I have two systems that were recently upgraded from Fedora 26 to 27 using DNF system-upgrade. Both were successfully running a KDE desktop for several generations previously, including on F26. In both cases, after the upgrade, starting KDE showed the splash screen and then a blank screen with only a pointer cursor. The keyboard and mouse worked but the screen remained blank. Both systems are set to start in systemd's multi-user mode; KDE is started with the startx command.
One of the systems is a 32-bit laptop with little local content. I chose to simply do a fresh install on this one to get it working quickly. Fedora doesn't offer an installation image for 32-bit F27 so I installed F25 and then did successive upgrades to F26 and then F27. KDE worked in both F25 and F26 but had the same problem once I got to F27. I did it again and stopped at F26 to have a working system.
The other system is a modern 64-bit desktop with 8GB RAM. This one still has the problem. I'd like to find the cause and a solution before I make the effort to revert to F26 because this one has a huge amount of local content which I would have to backup and restore. (I do backups but more for recovery of individual files; a general recovery could have other problems.)
Note that Gnome works on this machine (but I hate Gnome).
The logs haven't given me any clue as to the cause of the problem. I have now run "startx &>logfile" hoping that might tell me something. After the blank screen appeared, I killed xinit and Xorg from a different console. The logfile is included below in the hope it might spark some insight from the group.
xauth: file /home/dave/.serverauth.1517 does not exist X.Org X Server 1.19.6 Release Date: 2017-12-20 X Protocol Version 11, Revision 0 Build Operating System: 4.14.11-300.fc27.x86_64 Current Operating System: Linux dc7800.compata.com 4.14.16-300.fc27.x86_64 #1 SMP Wed Jan 31 19:24:27 UTC 2018 x86_64 Kernel command line: BOOT_IMAGE=/vmlinuz-4.14.16-300.fc27.x86_64 root=/dev/mapper/fedora_dc7800-root ro rd.lvm.lv=fedora_dc7800/root rd.lvm.lv=fedora_dc7800/swap vconsole.font=latarcyrheb-sun16 biosdevname=0 net.ifnames=0 selinux=0 LANG=en_US.UTF-8 Build Date: 10 January 2018 11:00:12PM Build ID: xorg-x11-server 1.19.6-2.fc27 Current version of pixman: 0.34.0 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/home/dave/.local/share/xorg/Xorg.0.log", Time: Tue Feb 6 21:11:34 2018 (==) Using config file: "/etc/X11/xorg.conf" (==) Using config directory: "/etc/X11/xorg.conf.d" (==) Using system config directory "/usr/share/X11/xorg.conf.d" startkde: Starting up... dbus-update-activation-environment: warning: error sending to systemd: org.freedesktop.DBus.Error.Spawn.ChildExited: Process org.freedesktop.systemd1 exited with status 1 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/akregator.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/blogilo.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/disable_kmix.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_akregator.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_global.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_global_toolbar.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/fonts_kate.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/importwizard.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kaddressbook.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kalarm.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/katomic-levelset.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kcalcrc.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kget.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kgthemeprovider-migration.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kleopatra.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/klickety.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kmail.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kmail2.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/knavalbattle.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/knotes.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/kontact.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/korganizer.upd' for new updates 2018-02-06T21:12:01 Checking update-file '/usr/share/kconf_update/krdb_libpathwipe.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/krunnerplugins.upd' for new updates 2018-02-06T21:12:02 krunnerplugins.upd: Found new update '5.9KRunnerPlugins' 2018-02-06T21:12:02 krunnerplugins.upd:3:'Script=krunnerplugins': Script 'krunnerplugins' not found 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/kscreenlocker.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/messageviewer.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/okular.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/pimsettingexporter.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/sieveeditor.upd' for new updates 2018-02-06T21:12:02 Checking update-file '/usr/share/kconf_update/webengineurlinterceptoradblock.upd' for new updates vmware-user: could not open /proc/fs/vmblock/dev Service started, version: 7.0.0 Service started, version: 7.0.0 Service started, version: 7.0.0 OpenGL vendor string: Intel Open Source Technology Center OpenGL renderer string: Mesa DRI Intel(R) Q35 OpenGL version string: 1.4 Mesa 17.2.4 OpenGL shading language version string: 1.20 Driver: Intel GPU class: i915/i945 OpenGL version: 1.4 GLSL version: 1.20 Mesa version: 17.2.4 Linux kernel version: 4.14.16 Requires strict binding: yes GLSL shaders: limited Texture NPOT support: yes Virtual Machine: no (II) AIGLX: Suspending AIGLX clients for VT switch (krb5-auth-dialog:2069): libnotify-WARNING **: Failed to connect to proxy (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Failed to read server caps (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Unsupported cache type for 'KEYRING:persistent:1000' (krb5-auth-dialog:2069): KrbAuthDialog-CRITICAL **: monitor_ccache: assertion 'ccache_name != NULL' failed Killed XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0" after 27 requests (27 known processed) with 0 events remaining. klauncher: Exiting on signal 1 Unexpected response from KInit (response = 0). startkde: Could not start ksmserver. Check your installation. -- I think this is where KDE hung and the following is after I killed it. Error: Can't open display: :0 KCrash: Application 'ksmserver' crashing... KCrash: Attempting to start /usr/libexec/drkonqi from kdeinit Warning: connect() failed: : No such file or directory KCrash: Attempting to start /usr/libexec/drkonqi directly Could not connect to D-Bus server: org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-SJIFqD8xwP: Connection refused startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
I'm guessing some dependent systemd service did not start causing the KDE startup issue. Instead of startx, install a login manager like KDM or SLIM and enable the system to launch GUI at boot and then reboot the system and see if the issue gets resolved.
-- Prasad
On Wed, 2018-02-07 at 11:24 +0530, Prasad K wrote:
startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
I'm guessing some dependent systemd service did not start causing the KDE startup issue. Instead of startx, install a login manager like KDM or SLIM and enable the system to launch GUI at boot and then reboot the system and see if the issue gets resolved.
The recommended login manager for KDE is SDDM. KDM also works but is older.
You might want to post on the Fedora KDE list for this kind of problem.
See https://lists.fedoraproject.org/admin/lists/kde.lists.fedoraproject.org/
poc
On 02/07/18 18:31, Patrick O'Callaghan wrote:
On Wed, 2018-02-07 at 11:24 +0530, Prasad K wrote:
startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
I'm guessing some dependent systemd service did not start causing the KDE startup issue. Instead of startx, install a login manager like KDM or SLIM and enable the system to launch GUI at boot and then reboot the system and see if the issue gets resolved.
The recommended login manager for KDE is SDDM. KDM also works but is older.
Right.
Yet, if it is an X-server problem neither will work. :-)
You might want to post on the Fedora KDE list for this kind of problem.
See https://lists.fedoraproject.org/admin/lists/kde.lists.fedoraproject.org/
On Wed, 2018-02-07 at 21:26 +0800, Ed Greshko wrote:
On 02/07/18 18:31, Patrick O'Callaghan wrote:
On Wed, 2018-02-07 at 11:24 +0530, Prasad K wrote:
startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
I'm guessing some dependent systemd service did not start causing the KDE startup issue. Instead of startx, install a login manager like KDM or SLIM and enable the system to launch GUI at boot and then reboot the system and see if the issue gets resolved.
The recommended login manager for KDE is SDDM. KDM also works but is older.
Right.
Yet, if it is an X-server problem neither will work. :-)
Indeed.
poc
"Patrick O'Callaghan" wrote:
You might want to post on the Fedora KDE list for this kind of problem.
I may try that. But the XIO errors and the failure to open the display made me think this might be a better place.
On Wed, 2018-02-07 at 21:46 -0800, Dave Close wrote:
"Patrick O'Callaghan" wrote:
You might want to post on the Fedora KDE list for this kind of problem.
I may try that. But the XIO errors and the failure to open the display made me think this might be a better place.
I think you said that Gnome does start but KDE doesn't, so it's likely that something in the KDE config is causing the problem.
poc
On 02/08/18 19:30, Patrick O'Callaghan wrote:
On Wed, 2018-02-07 at 21:46 -0800, Dave Close wrote:
"Patrick O'Callaghan" wrote:
You might want to post on the Fedora KDE list for this kind of problem.
I may try that. But the XIO errors and the failure to open the display made me think this might be a better place.
I think you said that Gnome does start but KDE doesn't, so it's likely that something in the KDE config is causing the problem.
Well, unless GNOME is using Wayland?
But, one should try creating another user and then try starting KDE on that new pristine account.
Ed Greshko wrote:
On 02/08/18 19:30, Patrick O'Callaghan wrote:
I think you said that Gnome does start but KDE doesn't, so it's likely that something in the KDE config is causing the problem.
Well, unless GNOME is using Wayland?
The way I read the release notes, it is. Though there apparently is a way to override and force it back to X. I haven't tried, mostly because I really don't like Gnome so I avoid it.
But, one should try creating another user and then try starting KDE on that new pristine account.
I've done that without any change.
On 02/08/18 23:57, Dave Close wrote:
Ed Greshko wrote:
On 02/08/18 19:30, Patrick O'Callaghan wrote:
I think you said that Gnome does start but KDE doesn't, so it's likely that something in the KDE config is causing the problem.
Well, unless GNOME is using Wayland?
The way I read the release notes, it is. Though there apparently is a way to override and force it back to X. I haven't tried, mostly because I really don't like Gnome so I avoid it.
But, one should try creating another user and then try starting KDE on that new pristine account.
I've done that without any change.
Hummm.... Very odd....
Looking at your Xorg.0.log I see this at the end.
[ 221.964] (II) AIGLX: Suspending AIGLX clients for VT switch [ 222.058] (II) systemd-logind: got pause for 13:65 [ 222.058] (II) systemd-logind: got pause for 13:69 [ 222.059] (II) systemd-logind: got pause for 13:64 [ 222.059] (II) systemd-logind: got pause for 226:0 [ 222.059] (II) systemd-logind: got pause for 13:66 [ 222.059] (II) systemd-logind: got pause for 13:67 [ 222.059] (II) systemd-logind: got pause for 13:68
You should check on your working system what you get. I don't have any of those types of pause messages.
These are the systemd-logind lines in my file on a VM.
[egreshko@f27k xorg]$ grep systemd-logi Xorg.0.log
[ 85.105] (II) systemd-logind: took control of session /org/freedesktop/login1/session/_31 [ 85.111] (II) systemd-logind: got fd for /dev/dri/card0 226:0 fd 11 paused 0 [ 85.644] (II) systemd-logind: got fd for /dev/input/event0 13:64 fd 19 paused 0 [ 85.650] (II) systemd-logind: got fd for /dev/input/event1 13:65 fd 22 paused 0 [ 85.651] (II) systemd-logind: got fd for /dev/input/event5 13:69 fd 23 paused 0 [ 85.658] (II) systemd-logind: got fd for /dev/input/event6 13:70 fd 24 paused 0 [ 85.713] (II) systemd-logind: got fd for /dev/input/event4 13:68 fd 25 paused 0 [ 85.723] (II) systemd-logind: got fd for /dev/input/event2 13:66 fd 26 paused 0 [ 85.729] (II) systemd-logind: got fd for /dev/input/event3 13:67 fd 27 paused 0 [ 4965.630] (II) systemd-logind: releasing fd for 13:67 [ 4965.936] (II) systemd-logind: releasing fd for 13:66 [ 4965.940] (II) systemd-logind: releasing fd for 13:68 [ 4966.008] (II) systemd-logind: releasing fd for 13:70 [ 4966.012] (II) systemd-logind: releasing fd for 13:69 [ 4966.015] (II) systemd-logind: releasing fd for 13:65 [ 4966.022] (II) systemd-logind: releasing fd for 13:64
I suppose you could try with selinux disabled as this appears to be a system-wide issue as opposed to a use configuration issue. I've upgraded several systems without issue. The HW in my case is all nVidia.
Probably should ask as well on the KDE list.
Ed Greshko wrote:
Looking at your Xorg.0.log I see this at the end.
[ 221.964] (II) AIGLX: Suspending AIGLX clients for VT switch [ 222.058] (II) systemd-logind: got pause for 13:65 [ 222.058] (II) systemd-logind: got pause for 13:69 [ 222.059] (II) systemd-logind: got pause for 13:64 [ 222.059] (II) systemd-logind: got pause for 226:0 [ 222.059] (II) systemd-logind: got pause for 13:66 [ 222.059] (II) systemd-logind: got pause for 13:67 [ 222.059] (II) systemd-logind: got pause for 13:68
You should check on your working system what you get. I don't have any of those types of pause messages.
I suppose you could try with selinux disabled as this appears to be a system-wide issue as opposed to a use configuration issue.
SElinux is disabled.
On the working system, the above messages are followed by
[ 72483.987] (II) AIGLX: Resuming AIGLX clients after VT switch
I'm not switching consoles. I haven't yet reached the opportunity to do so when the system hangs and the black screen appears. After that, yes, I can and do switch to a text console so I can kill the hung KDE session.
Searching for references to that AIGLX message, I find several reports, some going back a few years. The solutions are things like installing a different GUI so they aren't very helpful.
But it certainly does look like the systemd-logind pause is stuck and could be the cause of my problem.
Have I mentioned that I installed TigerVNC on the system, then connected from a different one? In that situation, KDE starts fine. The problem is only on the main screen.
I wrote:
But it certainly does look like the systemd-logind pause is stuck and could be the cause of my problem.
BTW, /etc/systemd/logind.conf contains only comments and none of the other logind configuration files are present.
On 02/09/18 13:26, Dave Close wrote:
Ed Greshko wrote:
Looking at your Xorg.0.log I see this at the end.
[ 221.964] (II) AIGLX: Suspending AIGLX clients for VT switch [ 222.058] (II) systemd-logind: got pause for 13:65 [ 222.058] (II) systemd-logind: got pause for 13:69 [ 222.059] (II) systemd-logind: got pause for 13:64 [ 222.059] (II) systemd-logind: got pause for 226:0 [ 222.059] (II) systemd-logind: got pause for 13:66 [ 222.059] (II) systemd-logind: got pause for 13:67 [ 222.059] (II) systemd-logind: got pause for 13:68
You should check on your working system what you get. I don't have any of those types of pause messages.
I suppose you could try with selinux disabled as this appears to be a system-wide issue as opposed to a use configuration issue.
SElinux is disabled.
On the working system, the above messages are followed by
[ 72483.987] (II) AIGLX: Resuming AIGLX clients after VT switch
I'm not switching consoles. I haven't yet reached the opportunity to do so when the system hangs and the black screen appears. After that, yes, I can and do switch to a text console so I can kill the hung KDE session.
Searching for references to that AIGLX message, I find several reports, some going back a few years. The solutions are things like installing a different GUI so they aren't very helpful.
But it certainly does look like the systemd-logind pause is stuck and could be the cause of my problem.
Have I mentioned that I installed TigerVNC on the system, then connected from a different one? In that situation, KDE starts fine. The problem is only on the main screen.
Is your system a desktop with a single monitor? A laptop with external monitor? Or something else?
If you do use either sddm or kdm do you get a login screen? Couldn't hurt to try that as Samuel has suggested.
Ed Greshko wrote:
Is your system a desktop with a single monitor? A laptop with external monitor? Or something else?
It's a desktop with a single monitor.
If you do use either sddm or kdm do you get a login screen? Couldn't hurt to try that as Samuel has suggested.
Using kdm, I get exactly the same result.
On 02/10/18 13:55, Dave Close wrote:
Ed Greshko wrote:
Is your system a desktop with a single monitor? A laptop with external monitor? Or something else?
It's a desktop with a single monitor.
If you do use either sddm or kdm do you get a login screen? Couldn't hurt to try that as Samuel has suggested.
Using kdm, I get exactly the same result.
Sorry, would you clarify?
You do get a login screen and then when you enter your PW you end up with a black screen with a cursor you can move around?
Ed Greshko wrote:
Is your system a desktop with a single monitor? A laptop with external monitor? Or something else?
It's a desktop with a single monitor.
If you do use either sddm or kdm do you get a login screen? Couldn't hurt to try that as Samuel has suggested.
Using kdm, I get exactly the same result.
Sorry, would you clarify?
You do get a login screen and then when you enter your PW you end up with a black screen with a cursor you can move around?
Yes. Exactly.
On 02/10/18 14:25, Dave Close wrote:
Ed Greshko wrote:
Is your system a desktop with a single monitor? A laptop with external monitor? Or something else?
It's a desktop with a single monitor.
If you do use either sddm or kdm do you get a login screen? Couldn't hurt to try that as Samuel has suggested.
Using kdm, I get exactly the same result.
Sorry, would you clarify?
You do get a login screen and then when you enter your PW you end up with a black screen with a cursor you can move around?
Yes. Exactly.
OK then at that point the X-server is functioning normally. Would it be possible to do the following?
1. Reboot the system to get to the login screen.
2. From a second system, ssh into the failing system, become root, and "tail -f /var/log/Xorg.0.log" to see what was last entered.
3. Login from the GUI on failing system and then check the tail to see if the "AIGLX: Suspending AIGLX clients for VT switch" appears at that point.
Ed Greshko wrote:
You do get a login screen and then when you enter your PW you end up with a black screen with a cursor you can move around?
Yes. Exactly.
OK then at that point the X-server is functioning normally. Would it be possible to do the following?
Reboot the system to get to the login screen.
From a second system, ssh into the failing system, become root,
and "tail -f /var/log/Xorg.0.log" to see what was last entered.
- Login from the GUI on failing system and then check the tail to
see if the "AIGLX: Suspending AIGLX clients for VT switch" appears at that point.
I did as described. In this case, the default target was graphical. The last thing in Xorg.0.log are messages about the keyboard hotkeys. After I entered my password, the KDE splash screen appeared with a working cursor. But instead of going to a black screen, it stayed on the splash screen. The Xorg.0.log did not change until I switched to VT2, then it showed both AIGLX suppended and resuming when I switched back to the hung VT1.
On 02/11/18 02:42, Dave Close wrote:
Ed Greshko wrote:
You do get a login screen and then when you enter your PW you end up with a black screen with a cursor you can move around?
Yes. Exactly.
OK then at that point the X-server is functioning normally. Would it be possible to do the following?
Reboot the system to get to the login screen.
From a second system, ssh into the failing system, become root,
and "tail -f /var/log/Xorg.0.log" to see what was last entered.
- Login from the GUI on failing system and then check the tail to
see if the "AIGLX: Suspending AIGLX clients for VT switch" appears at that point.
I did as described. In this case, the default target was graphical. The last thing in Xorg.0.log are messages about the keyboard hotkeys. After I entered my password, the KDE splash screen appeared with a working cursor. But instead of going to a black screen, it stayed on the splash screen. The Xorg.0.log did not change until I switched to VT2, then it showed both AIGLX suppended and resuming when I switched back to the hung VT1.
OK. So, those lines are pretty much meaningless for the problem.
Can you run "dnf distro-sync" just to make sure all packages are at the most current level?
Ed Greshko wrote:
Can you run "dnf distro-sync" just to make sure all packages are at the most current level?
Dependencies resolved. Nothing to do.
On 02/11/18 03:56, Dave Close wrote:
Ed Greshko wrote:
Can you run "dnf distro-sync" just to make sure all packages are at the most current level?
Dependencies resolved. Nothing to do.
OK....
If you ssh into the failing system and run "top" does anything seem odd? A process stuck at the top or hogging CPU?
Ed Greshko wrote:
If you ssh into the failing system and run "top" does anything seem odd? A process stuck at the top or hogging CPU?
99.7% idle. Occasionally, plasmashell or vncserver will show a few percent. Usually top itself is the big user.
On 02/11/18 04:21, Dave Close wrote:
Ed Greshko wrote:
If you ssh into the failing system and run "top" does anything seem odd? A process stuck at the top or hogging CPU?
99.7% idle. Occasionally, plasmashell or vncserver will show a few percent. Usually top itself is the big user.
Grasping at straws here....
Are you running vnc server as a systemd service that gets started at boot time? I've never had a problem with that....but if you're doing that how about disabling it for a test.
Ed Greshko wrote:
Grasping at straws here....
Are you running vnc server as a systemd service that gets started at boot time? I've never had a problem with that....but if you're doing that how about disabling it for a test.
No, I started it from the command line. Recall that I had the same problem before I started using VNC. In fact, VNC is just a workaround so I can get practical work done using this machine.
I have mixed feelings about your frustration. I'm certainly frustrated myself. But it feels slightly good that someone else is also baffled by the situation. I only wish there were someone who wasn't baffled and could tell me what is wrong. Thanks for all the help.
On 02/11/18 04:56, Dave Close wrote:
Ed Greshko wrote:
Grasping at straws here....
Are you running vnc server as a systemd service that gets started at boot time? I've never had a problem with that....but if you're doing that how about disabling it for a test.
No, I started it from the command line. Recall that I had the same problem before I started using VNC. In fact, VNC is just a workaround so I can get practical work done using this machine.
I have mixed feelings about your frustration. I'm certainly frustrated myself. But it feels slightly good that someone else is also baffled by the situation. I only wish there were someone who wasn't baffled and could tell me what is wrong. Thanks for all the help.
Well, you have plasmashell running so it is a mystery as to what may be going on. Just after login on a system I see...
[egreshko@acer ~]$ ps -eaf | grep plasma egreshko 1561 1485 9 06:44 ? 00:00:07 /usr/bin/plasmashell egreshko 1816 1468 0 06:44 ? 00:00:00 desktop.so [kdeinit5] desktop local:/run/user/1029/klauncherTJ1469.1.slave-socket local:/run/user/1029/plasmashellTJ1561.1.slave-socket egreshko 1873 1468 0 06:44 ? 00:00:00 file.so [kdeinit5] file local:/run/user/1029/klauncherTJ1469.1.slave-socket local:/run/user/1029/plasmashellXM1561.2.slave-socket egreshko 1874 1468 0 06:44 ? 00:00:00 trash.so [kdeinit5] trash local:/run/user/1029/klauncherTJ1469.1.slave-socket local:/run/user/1029/plasmashellLh1561.3.slave-socket egreshko 1876 1468 0 06:44 ? 00:00:00 file.so [kdeinit5] file local:/run/user/1029/klauncherTJ1469.1.slave-socket local:/run/user/1029/plasmashellnS1561.4.slave-socket egreshko 1877 1468 0 06:44 ? 00:00:00 trash.so [kdeinit5] trash local:/run/user/1029/klauncherTJ1469.1.slave-socket local:/run/user/1029/plasmashellkn1561.5.slave-socket
I don't know of a way to restart plasmashell remotely....
So, I suppose I try strace attached to the plasmashell PID and see if anything pops out.
And I'd probably post the current status and findings to the KDE list to see if someone there has an idea. And, of course, you may want to consider a bugzilla.
Oh, one thing has just come to mind....
This is a system which has been upgraded from F26 to F27, right? So, it still has an F26 kernel on it. So, does KDE work if you boot the F26 kernel?
On 02/08/2018 09:26 PM, Dave Close wrote:
Ed Greshko wrote:
Looking at your Xorg.0.log I see this at the end.
[ 221.964] (II) AIGLX: Suspending AIGLX clients for VT switch [ 222.058] (II) systemd-logind: got pause for 13:65 [ 222.058] (II) systemd-logind: got pause for 13:69 [ 222.059] (II) systemd-logind: got pause for 13:64 [ 222.059] (II) systemd-logind: got pause for 226:0 [ 222.059] (II) systemd-logind: got pause for 13:66 [ 222.059] (II) systemd-logind: got pause for 13:67 [ 222.059] (II) systemd-logind: got pause for 13:68
You should check on your working system what you get. I don't have any of those types of pause messages.
That's because you aren't switching to a different console to read the log.
On the working system, the above messages are followed by
[ 72483.987] (II) AIGLX: Resuming AIGLX clients after VT switch
I'm not switching consoles. I haven't yet reached the opportunity to do so when the system hangs and the black screen appears. After that, yes, I can and do switch to a text console so I can kill the hung KDE session.
You do switch consoles and you just again said you did. You switched consoles and then killed the session. I get those same messages when I switch to a text console and then back again. Those messages are normal and not a problem.
But it certainly does look like the systemd-logind pause is stuck and could be the cause of my problem.
Try switching back to the "stuck" X session, wait 10 seconds, and then switch back again to the console. See what is in the log. I bet you will see the logind resume and the AIGLX resume messages.
Have I mentioned that I installed TigerVNC on the system, then connected from a different one? In that situation, KDE starts fine. The problem is only on the main screen.
This starts the session differently. How are you starting vnc and KDE in this case. That might give a hint to why the other method isn't working.
Samuel Sieb wrote:
I'm not switching consoles. I haven't yet reached the opportunity to do so when the system hangs and the black screen appears. After that, yes, I can and do switch to a text console so I can kill the hung KDE session.
You do switch consoles and you just again said you did. You switched consoles and then killed the session. I get those same messages when I switch to a text console and then back again. Those messages are normal and not a problem.
I ran the test this evening while I was remotely connected to the machine from a different machine. When KDE stuck, I checked the Xorg log from the other machine, without switching consoles on the machine with the problem. The log ended like this:
[261624.860] (II) AIGLX: Suspending AIGLX clients for VT switch [261624.925] (II) systemd-logind: got pause for 13:65 [261624.925] (II) systemd-logind: got pause for 13:69 [261624.925] (II) systemd-logind: got pause for 226:0 [261624.925] (II) systemd-logind: got pause for 13:64 [261624.925] (II) systemd-logind: got pause for 13:66 [261624.926] (II) systemd-logind: got pause for 13:68 [261624.926] (II) systemd-logind: got pause for 13:67
So, no, I hadn't tried to switch consoles when those messages appeared.
Have I mentioned that I installed TigerVNC on the system, then connected from a different one? In that situation, KDE starts fine. The problem is only on the main screen.
This starts the session differently. How are you starting vnc and KDE in this case. That might give a hint to why the other method isn't working.
Well, I'm certainly seeking hints. The VNC session got me to a blank screen image. I opened an xterm on it and ran startx.
On 02/09/2018 10:04 PM, Dave Close wrote:
Samuel Sieb wrote:
I'm not switching consoles. I haven't yet reached the opportunity to do so when the system hangs and the black screen appears. After that, yes, I can and do switch to a text console so I can kill the hung KDE session.
You do switch consoles and you just again said you did. You switched consoles and then killed the session. I get those same messages when I switch to a text console and then back again. Those messages are normal and not a problem.
I ran the test this evening while I was remotely connected to the machine from a different machine. When KDE stuck, I checked the Xorg log from the other machine, without switching consoles on the machine with the problem. The log ended like this:
[261624.860] (II) AIGLX: Suspending AIGLX clients for VT switch [261624.925] (II) systemd-logind: got pause for 13:65 [261624.925] (II) systemd-logind: got pause for 13:69 [261624.925] (II) systemd-logind: got pause for 226:0 [261624.925] (II) systemd-logind: got pause for 13:64 [261624.925] (II) systemd-logind: got pause for 13:66 [261624.926] (II) systemd-logind: got pause for 13:68 [261624.926] (II) systemd-logind: got pause for 13:67
So, no, I hadn't tried to switch consoles when those messages appeared.
That's strange because those lines are only caused by a VT switch as it says...
Have I mentioned that I installed TigerVNC on the system, then connected from a different one? In that situation, KDE starts fine. The problem is only on the main screen.
This starts the session differently. How are you starting vnc and KDE in this case. That might give a hint to why the other method isn't working.
Well, I'm certainly seeking hints. The VNC session got me to a blank screen image. I opened an xterm on it and ran startx.
You what? You ran startx from within an X session?
Samuel Sieb wrote:
Well, I'm certainly seeking hints. The VNC session got me to a blank screen image. I opened an xterm on it and ran startx.
You what? You ran startx from within an X session?
Well, it's been a few days. I agree that doesn't make sense. I'll try it again and post a clarification.
On 02/09/2018 10:27 PM, Dave Close wrote:
Samuel Sieb wrote:
Well, I'm certainly seeking hints. The VNC session got me to a blank screen image. I opened an xterm on it and ran startx.
You what? You ran startx from within an X session?
Well, it's been a few days. I agree that doesn't make sense. I'll try it again and post a clarification.
Also try running "systemctl isolate graphical".
Samuel Sieb wrote:
Well, I'm certainly seeking hints. The VNC session got me to a blank screen image. I opened an xterm on it and ran startx.
You what? You ran startx from within an X session?
Well, it's been a few days. I agree that doesn't make sense. I'll try it again and post a clarification.
Sorry about the confusion. There was indeed an xterm involved but it was where I started the VNC session; it was running on the remote deskop. VNC started directly in KDE as expected.
Also try running "systemctl isolate graphical".
I generally run two VTs, VT1 is a text root console, VT2 is where I run startx under my own account. In this case, startx was hung on VT2 as I have previously described. I ran the isolate command from a remote login. The main screen immediately switched to a sddm login -- running on VT1. Each time I entered my password, it cleared for a moment, back to the root text console, then reappeared asking for password again. Even killing Xorg and sddm via the remote login did not terminate it. Meanwhile, the hung KDE on VT2 was unchanged.
The sddm login initially showed Gnome as the default desktop, though it never got there. Changing the selection to Plasma changed the behavior slightly. Then after entering my password, nothing happened. Though the cursor continued to work, no selection had any effect. I rebooted to clear it.
During this experiment, /etc/sysconfig/desktop contained,
DESKTOP="KDE" DISPLAYMANAGER="KDE"
I wrote:
I have two systems that were recently upgraded from Fedora 26 to 27 using DNF system-upgrade. Both were successfully running a KDE desktop for several generations previously, including on F26. In both cases, after the upgrade, starting KDE showed the splash screen and then a blank screen with only a pointer cursor. The keyboard and mouse worked but the screen remained blank. Both systems are set to start in systemd's multi-user mode; KDE is started with the startx command.
Prasad K wrote:
I'm guessing some dependent systemd service did not start causing the KDE startup issue. Instead of startx, install a login manager like KDM or SLIM and enable the system to launch GUI at boot and then reboot the system and see if the issue gets resolved.
I tried that with the first system, the one which I have already re-installed, before the reinstall. The result was the same except that killing the hung system was more painful.
On 02/07/18 13:27, Dave Close wrote:
(==) Using config file: "/etc/X11/xorg.conf"
First, check the file /home/dave/.local/share/xorg/Xorg.0.log for the complete Xorg log. Looking for EE lines
On all of my systems that file doesn't exist since everything is determined by the kernel. That line doesn't even appear in my Xorg.0.log. So, maybe try starting without that file in place. What does it contain?
Ed Greshko wrote:
(==) Using config file: "/etc/X11/xorg.conf"
First, check the file /home/dave/.local/share/xorg/Xorg.0.log for the complete Xorg log. Looking for EE lines
I found only two such lines, each present only in one of eight logs.
(1) intel(0): Failed to submit rendering commands (Invalid argument), disabling acceleration.
(2) open /dev/fb0: Permission denied
For (2), I tried chmod 666 on the device but that didn't help, except that the EE line is no longer present. For (1), if acceleration were the only problem, I think I'd see a screen.
On all of my systems that file doesn't exist since everything is determined by the kernel. That line doesn't even appear in my Xorg.0.log. So, maybe try starting without that file in place. What does it contain?
I moved /etc/X11/xorg.conf and /etc/X11/xorg.conf.d to a different directory and tried again. No change. The Xorg log file is basically the same as before, no EE lines.
The xorg.conf file looks about the way I would expect. Frankly, it looks pretty vanilla. The xorg.conf.d directory contains only a 00-keyboard.conf file.
On 02/08/18 13:40, Dave Close wrote:
Ed Greshko wrote:
(==) Using config file: "/etc/X11/xorg.conf"
First, check the file /home/dave/.local/share/xorg/Xorg.0.log for the complete Xorg log. Looking for EE lines
I found only two such lines, each present only in one of eight logs.
(1) intel(0): Failed to submit rendering commands (Invalid argument), disabling acceleration.
(2) open /dev/fb0: Permission denied
For (2), I tried chmod 666 on the device but that didn't help, except that the EE line is no longer present. For (1), if acceleration were the only problem, I think I'd see a screen.
I'm not very familiar with the intel driver. It may not even use /dev/fb0.
On all of my systems that file doesn't exist since everything is determined by the kernel. That line doesn't even appear in my Xorg.0.log. So, maybe try starting without that file in place. What does it contain?
I moved /etc/X11/xorg.conf and /etc/X11/xorg.conf.d to a different directory and tried again. No change. The Xorg log file is basically the same as before, no EE lines.
The xorg.conf file looks about the way I would expect. Frankly, it looks pretty vanilla. The xorg.conf.d directory contains only a 00-keyboard.conf file.
Could you upload /home/dave/.local/share/xorg/Xorg.0.log for others to look at? The one on my test system is 402 lines long so yours may be equally long.
Ed Greshko wrote:
Could you upload /home/dave/.local/share/xorg/Xorg.0.log for others to look at? The one on my test system is 402 lines long so yours may be equally long.
Sure. Here's the latest version.
On 02/06/2018 09:27 PM, Dave Close wrote:
One of the systems is a 32-bit laptop with little local content. I chose to simply do a fresh install on this one to get it working quickly. Fedora doesn't offer an installation image for 32-bit F27 so I installed F25 and then did successive upgrades to F26 and then F27. KDE worked in both F25 and F26 but had the same problem once
There is no live image for 32-bit, but there is still an "everything" netinstall image.
On 02/06/2018 09:27 PM, Dave Close wrote:
I have two systems that were recently upgraded from Fedora 26 to 27 using DNF system-upgrade. Both were successfully running a KDE desktop for several generations previously, including on F26. In both cases, after the upgrade, starting KDE showed the splash screen and then a blank screen with only a pointer cursor. The keyboard and mouse worked but the screen remained blank. Both systems are set to start in systemd's multi-user mode; KDE is started with the startx command.
This is likely a problem. Why don't you use the graphical target mode? I expect that the xinit is not starting everything that needs to be running. You could try running "systemctl start sddm" to start the login manager.
The logs haven't given me any clue as to the cause of the problem. I have now run "startx &>logfile" hoping that might tell me something. After the blank screen appeared, I killed xinit and Xorg from a different console. The logfile is included below in the hope it might spark some insight from the group.
xauth: file /home/dave/.serverauth.1517 does not exist X.Org X Server 1.19.6 Release Date: 2017-12-20 X Protocol Version 11, Revision 0 Build Operating System: 4.14.11-300.fc27.x86_64 Current Operating System: Linux dc7800.compata.com 4.14.16-300.fc27.x86_64 #1 SMP Wed Jan 31 19:24:27 UTC 2018 x86_64 Kernel command line: BOOT_IMAGE=/vmlinuz-4.14.16-300.fc27.x86_64 root=/dev/mapper/fedora_dc7800-root ro rd.lvm.lv=fedora_dc7800/root rd.lvm.lv=fedora_dc7800/swap vconsole.font=latarcyrheb-sun16 biosdevname=0 net.ifnames=0 selinux=0 LANG=en_US.UTF-8 Build Date: 10 January 2018 11:00:12PM Build ID: xorg-x11-server 1.19.6-2.fc27 Current version of pixman: 0.34.0 Before reporting problems, check http://wiki.x.org to make sure that you have the latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, (II) informational, (WW) warning, (EE) error, (NI) not implemented, (??) unknown. (==) Log file: "/home/dave/.local/share/xorg/Xorg.0.log", Time: Tue Feb 6 21:11:34 2018 (==) Using config file: "/etc/X11/xorg.conf" (==) Using config directory: "/etc/X11/xorg.conf.d" (==) Using system config directory "/usr/share/X11/xorg.conf.d"
At this point X is running.
startkde: Starting up... dbus-update-activation-environment: warning: error sending to systemd: org.freedesktop.DBus.Error.Spawn.ChildExited: Process org.freedesktop.systemd1 exited with status 1
Potentially an issue, I don't know what it's trying to do.
OpenGL vendor string: Intel Open Source Technology Center OpenGL renderer string: Mesa DRI Intel(R) Q35 OpenGL version string: 1.4 Mesa 17.2.4 OpenGL shading language version string: 1.20 Driver: Intel GPU class: i915/i945 OpenGL version: 1.4 GLSL version: 1.20 Mesa version: 17.2.4 Linux kernel version: 4.14.16 Requires strict binding: yes GLSL shaders: limited Texture NPOT support: yes Virtual Machine: no
Maybe the window manager starting up.
(II) AIGLX: Suspending AIGLX clients for VT switch
This is where you switched to another console.
(krb5-auth-dialog:2069): libnotify-WARNING **: Failed to connect to proxy (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Failed to read server caps (krb5-auth-dialog:2069): KrbAuthDialog-WARNING **: Unsupported cache type for 'KEYRING:persistent:1000' (krb5-auth-dialog:2069): KrbAuthDialog-CRITICAL **: monitor_ccache: assertion 'ccache_name != NULL' failed
With no timestamps it's hard to tell if this happened before you killed things. Are you using Kerberos on the system?
Killed XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0" after 27 requests (27 known processed) with 0 events remaining.
You killed the X server.
klauncher: Exiting on signal 1 Unexpected response from KInit (response = 0). startkde: Could not start ksmserver. Check your installation. -- I think this is where KDE hung and the following is after I killed it. Error: Can't open display: :0 KCrash: Application 'ksmserver' crashing... KCrash: Attempting to start /usr/libexec/drkonqi from kdeinit Warning: connect() failed: : No such file or directory KCrash: Attempting to start /usr/libexec/drkonqi directly Could not connect to D-Bus server: org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-SJIFqD8xwP: Connection refused startkde: Shutting down... kdeinit5_wrapper: Warning: connect(/run/user/1000/kdeinit5__0) failed: : No such file or directory Error: Can not contact kdeinit5! xprop: unable to open display ':0' xprop: unable to open display ':0' startkde: Done.
And everything else fails after that.
I'm pretty sure you're just missing some piece because of how you're starting it. Gnome and KDE are really not designed to be started this way.
On 02/09/18 12:37, Samuel Sieb wrote:
I'm pretty sure you're just missing some piece because of how you're starting it. Gnome and KDE are really not designed to be started this way.
FWIW, KDE starts just fine for me on several F27 systems using "startx".