| Summary: | Logging into Xfce with lightdm impossible, but logging into IceWM works | ||
|---|---|---|---|
| Product: | Mageia | Reporter: | magnux77 <magnux77> |
| Component: | RPM Packages | Assignee: | Jani Välimaa <jani.valimaa> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Normal | CC: | bittwister2, mageia, magnux77, marja11 |
| Version: | Cauldron | Keywords: | 7beta2 |
| Target Milestone: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Source RPM: | CVE: | ||
| Status comment: | |||
| Attachments: |
/root/drakx/report.bug.xz
.xsession-errors XFCE Xorg0.log journal.log hardware description |
||
|
Description
magnux77
2019-01-17 15:40:53 CET
seen in journalctl : -- Le chargement de l'espace utilisateur a nécessité 37174757 microsecondes. janv. 16 02:27:25 localhost lightdm[2090]: pam_succeed_if(lightdm:auth): requirement "user ingroup nopasswdlogin" not met by user "jps" janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1645 (lightdm) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1792 (lightdm-gtk3-gr) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1865 (dbus-launch) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1866 (dbus-daemon) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1868 (at-spi-bus-laun) with signal SIGTERM. janv. 16 02:27:31 localhost xfconfd[1872]: Name org.xfce.Xfconf lost on the message dbus, exiting. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1872 (xfconfd) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1878 (dbus-daemon) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: session-c1.scope: Killing process 1880 (at-spi2-registr) with signal SIGTERM. janv. 16 02:27:31 localhost systemd[1]: Stopping Session c1 of user lightdm. my password is correct, i tested it in a console. Assigning to the registered maintainer. Assignee:
bugsquad =>
jani.valimaa Can't reproduce. Is there something 'interesting' in lightdm logs? /var/log/lightdm/lightdm.log /var/log/lightdm/seat0-greeter.log /var/log/lightdm/x-0.log Check also ~/.xsession-errors after failed login. I forgot to say that i have had the same problem on another PC with M732 Xfce. Tried a login with Xfce that failed. Then in console, i moved the files you have asked for in my tmp folder. And then i examine them as i was logged within IceWM. I have only found errors in .xsession-errors. But i post (part of) the other files for advice. ---.xsession-errors---------------------- ** (xfce4-session:2597): CRITICAL **: 15:11:32.962: polkit_unix_process_set_property: assertion 'val != -1' failed (xfce4-session:2597): xfce4-session-WARNING **: 15:11:33.104: xfsm_manager_load_session: Something wrong with /home/jps/.cache/sessions/xfce4-session-localhost:0, Does it exist? Permissions issue? gnome-keyring-daemon: insufficient process capabilities, insecure memory might get used (polkit-mate-authentication-agent-1:2669): GLib-CRITICAL **: 15:11:34.519: g_variant_new_string: assertion 'string != NULL' failed (polkit-mate-authentication-agent-1:2669): polkit-mate-1-WARNING **: 15:11:34.521: Failed to register client: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.gnome.SessionManager was not provided by any .service files Subroutine Gtk3::main redefined at /usr/share/perl5/vendor_perl/Gtk3.pm line 525. Subroutine Gtk3::main redefined at /usr/share/perl5/vendor_perl/Gtk3.pm line 525. (wrapper-2.0:2715): dbind-WARNING **: 15:11:40.398: Could not open X display Unable to init server: Impossible de se connecter : Connexion refusée (wrapper-2.0:2725): Gtk-WARNING **: 15:11:40.413: cannot open display: :0.0 Gdk-Message: 15:11:41.535: net_applet: Fatal IO error 2 (Aucun fichier ou dossier de ce type) on X server :0.0. Gdk-Message: 15:11:41.535: xfsettingsd: Fatal IO error 0 (Succès) on X server :0.0. Gdk-Message: 15:11:41.539: Thunar: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.0. light-locker: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.0. Gdk-Message: 15:11:41.541: polkit-mate-authentication-agent-1: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.0. Gdk-Message: 15:11:41.541: xfce4-notifyd: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.0. Gdk-Message: 15:11:41.541: xfwm4: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0.0. Gdk-Message: 15:11:41.545: xfce4-panel: Fatal IO error 4 (Appel système interrompu) on X server :0.0. Gdk-Message: 15:11:41.545: xfce4-session: Fatal IO error 11 (Ressource temporairement non disponible) on X server :0. Gdk-Message: 15:11:41.574: wrapper-2.0: Fatal IO error 0 (Succès) on X server :0.0. ----------------------------------------- x-0.log raises only warnings ---.x-0.log------------------------------ The XKEYBOARD keymap compiler (xkbcomp) reports: > Warning: Unsupported high keycode 372 for name <I372> ignored > X11 cannot support keycodes above 255. > This warning only shows for the first high keycode. Errors from xkbcomp are not fatal to the X server The XKEYBOARD keymap compiler (xkbcomp) reports: > Warning: Unsupported high keycode 372 for name <I372> ignored > X11 cannot support keycodes above 255. > This warning only shows for the first high keycode. Errors from xkbcomp are not fatal to the X server (II) AIGLX: Suspending AIGLX clients for VT switch ----------------------------------------- I did not see any errors in lightdm.log. Here is the only end of that log for you to see. ---lightdm.log--------------------------- [+147.70s] DEBUG: Launching process 2742: /usr/share/X11/xdm/Xsetup_0 [+147.95s] DEBUG: Process 2742 exited with return value 0 [+147.95s] DEBUG: Seat seat0: Exit status of /usr/share/X11/xdm/Xsetup_0: 0 [+147.95s] DEBUG: Session pid=2739: Running command /usr/sbin/lightdm-gtk3-greeter [+147.95s] DEBUG: Creating shared data directory /var/lib/lightdm-data/lightdm [+147.95s] DEBUG: Session pid=2739: Logging to /var/log/lightdm/seat0-greeter.log [+148.04s] DEBUG: Activating VT 1 [+148.04s] DEBUG: Activating login1 session c4 [+148.04s] DEBUG: Seat seat0 changes active session to c4 [+148.04s] DEBUG: Session c4 is already active [+148.52s] DEBUG: Greeter connected version=1.26.0 api=1 resettable=false [+149.14s] DEBUG: Greeter start authentication for jps [+149.14s] DEBUG: Session pid=2793: Started with service 'lightdm', username 'jps' [+149.15s] DEBUG: Session pid=2793: Got 1 message(s) from PAM [+149.15s] DEBUG: Prompt greeter with 1 message(s) [+150.82s] DEBUG: Seat seat0 changes active session to 5 ----------------------------------------- Just warnings in the short seat0-greeter.log ---seat0-greeter.log--------------------- ** Message: 15:11:43.838: Starting lightdm-gtk-greeter 2.0.6 (Dec 15 2018, 11:45:53) ** Message: 15:11:43.840: [Configuration] Reading file: /etc/lightdm/lightdm-gtk-greeter.conf.d/50-mageia.conf ** Message: 15:11:43.840: [Configuration] Reading file: /etc/lightdm/lightdm-gtk-greeter.conf (lightdm-gtk3-greeter:2763): Gtk-WARNING **: 15:11:44.906: Drawing a gadget with negative dimensions. Did you forget to allocate a size? (node menubar owner GreeterMenuBar) (lightdm-gtk3-greeter:2763): Gtk-WARNING **: 15:11:45.482: Drawing a gadget with negative dimensions. Did you forget to allocate a size? (node menubar owner GreeterMenuBar) -----------------------------------------
magnux77
2019-01-23 14:40:52 CET
CC:
(none) =>
magnux77 Beta round 3 The situation is still the same, nothing has changed. I worry. Xfce login impossible and IceWM windows buttons without symbol... Logs asked are still available above.
Running xfce in 64 bit hardware on three different nodes without problems.
I know for a fact that running su instead of su - root
can cause problems. Verify that all files are owned by you. Run
find $HOME \( -not -user $USER -or -not -group $USER \) -exec ls -al '{}' \;
Do not know about xfce, but I have seen the corrupted environment get saved
during log out, So the following might not work. Go ahead and remove
all /home/jps/.cache files, log out/in
I always do clean installs and always create two user accounts, normal and junk.
normal is used without changes to verify DM/DE is working system wide.
junk is the one I use to login into, run "su - root" and go clean out
my account .cache whenever I think there is Display Environment problem.
I also use junk for testing. I can delete/create it and start testing again
without dinking up my account.
I can suggest that you create a junk account and log into it. If it has
no problems you know it is not a system problem but a user problem.
Also, anytime there is a xfce-config updated package you are supposed
to remove ~/.config/xfce4 and ~/.config/Thunar and go back through setting
up xfce as you had it before. Try
cat /usr/share/doc/mageia-xfce-config/README.update.urpmiCC:
(none) =>
bittwister2 0° Thanks for your answer. It's good news to know Xfce is running on others hardware ;-)
1° Something I forgot to mention, it is a test PC. So I tried a second install with format sda3 (/home) : same problem.
2° I know that workaround of deleting files under /home/jps/.cache/sessions. But there is absolutely no file under sessions.
[jps@localhost sessions]$ ll -a
total 8
drwx------ 2 jps jps 4096 févr. 17 15:22 ./
drwxr-xr-x 7 jps jps 4096 févr. 17 16:15 ../
|jps@localhost sessions]$
3° I did create another user that could not log in too.
4° No files not owned.
[jps@localhost jps]$ find $HOME \( -not -user $USER -or -not -group $USER \) -exec ls -al '{}' \;
[jps@localhost jps]$
5° README.update.urpmi
As said above, /home has been formatted.
Might not hurt to provide some basic information like name of iso used for install, and results from the following commands. cat /etc/sysconfig/desktop systemctl status display-manager.service | grep -E "PID:|bin/" Might also help to attach the report.bug.xz file found in /root/drakx. The IceWM buttons problem is bug 24199. CC:
(none) =>
mageia Created attachment 10755 [details] /root/drakx/report.bug.xz Asked by Martin Whitaker in comment#9 1° Iso : Mageia-7-beta2-x86_64.iso
2° [jps@localhost ~]$ cat /etc/sysconfig/desktop
DISPLAYMANAGER=lightdm
[jps@localhost ~]$
3° [jps@localhost ~]$ systemctl status display-manager.service | grep -E "PID:|bin/"
Main PID: 1486 (lightdm)
├─1486 /usr/sbin/lightdm
[jps@localhost ~]$
4° /root/drakx/report.bug.xz
Thanks for your attention.
I've looked through the installer report, but can't see anything amiss there. I've done quite a few Xfce installs, and never seen anything like this. Either did i, in the different logs i discovered under /root/drakx. Didn't find anything. I tried another test on that PC, an install from a Live with Mageia-7-beta2-Live-Xfce-x86_64.iso. The problem is the same : after Live has booted, at lightdm panel, it is impossible to connect with the login "Live", lightdm panel disappears and then reappears. When logging with root ligthdm panel disappears, then a red screen appears for a second, then lightdm panel reappears. Created attachment 10771 [details]
.xsession-errors XFCE
Tried to focus on xfce start with .xsession-errors of just ONE try to connect.
I use Mageia-7-beta2-x86_64.iso installed that can log into IceWM but not Xfce and /home formatted.
1° At lightdm panel, i switch to a terminal and delete /home/jps/.xsession-errors.
2° In the lightdm panel i try to log in with Xfce (that fails).
3° In the terminal i "mv .xsession-errors XFCE.xsession-errors".
Then, back to lightdm panel, i log with IceWM to post that file. It seems to me that there are too many errors, even if i can not understand...
Can you repeat that test, but at step 3, capture a copy of /var/log/Xorg.0.log and the output of 'systemctl -ab > journal.log' (run as root). Created attachment 10773 [details]
Xorg0.log
Before log in, rm /var/log/Xorg.0.log, then log in XFCE, then save this Xorg.0.log (before log in with IceWM)
Created attachment 10774 [details]
journal.log
After log in Xfce fails, by terminal, not (systemctl but) journalctl -ab > journal.log
Thanks for your determination.
(In reply to magnux77 from comment #17) > Created attachment 10774 [details] > journal.log > > After log in Xfce fails, by terminal, not (systemctl but) journalctl -ab > > journal.log > > Thanks for your determination. The following line gives a hint that this might have something to do with X11 and/or video driver. févr. 20 17:21:41 localhost org.a11y.Bus[2029]: XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0" Seems I failed to save my last comment :-( Yes, there's definitely a problem connecting to the X server (lots more errors show up in .xsession-errors), but there's nothing in Xorg.0.log to indicate a problem, and both LightDM and IceWM are managing to connect. That suggests an X authentication problem to me - but I can't see any reason for it. I've installed Xfce on four different platforms now, and none of them have this problem. The only unusual message I can see in the journal is févr. 20 17:21:41 localhost kernel: lightdm[2158]: segfault at 8 ip 00007fad5fa5b856 sp 00007ffebb032e58 error 4 in libc-2.29.so[7fad5f9e2000+155000] I pushed a new lightdm build to BS with gcc 8.3.0, just in case. T1his problem occurs on a test PC that is not usefull for me. But i wonder if it can happens to anyone else. "there's definitely a problem connecting to the X server" I have also seen those messages. But i just don't know what to do to go further. I made all the updates, i have seen lightdm, but no way. I try a reinstall with format /home, no way. What are the specs of the machine? CPU, GPU, etc? Created attachment 10816 [details]
hardware description
Here is hardware description from "hwinfo --short > hardware.txt"
(In reply to magnux77 from comment #23) > Created attachment 10816 [details] > hardware description > > Here is hardware description from "hwinfo --short > hardware.txt" Which graphics card this machine has? Attached file only says: graphics card: Hewlett-Packard Company Presario V6133CL CCM says it's a : NVidia Corporation C51 [GeForce Go 6150] VGA compatible controller. My others PC run fine in M7. I decided to have another try on this PC. In fact, i made a lot of install with errors. I tried M7x86_64-Live-Xfce. No login possible. When trying user root, the screen become red but no login, immediatly back on the log in screen. I tried M7x86_64-Xfce and after login, i saw the xfce panel at the top of the screen, then the screen blinked during a few seconds and after it quietened. I don't know why, does anybody know ? Don't think so, but it is OK. Thanks, i close the bug Resolution:
(none) =>
FIXED |