lightdm not visible upon reboot

Bug #1254880 reported by gregrwm
18
This bug affects 4 people
Affects Status Importance Assigned to Milestone
lightdm (Ubuntu)
Fix Released
High
Unassigned

Bug Description

lightdm did not appear to come up upon reboot, tho these processes exist:

*j101 root 1600 -1 1600 1 1600 11:56:00 0 poll_s Ssl lightdm
*j102 root 1665 1665 1665 1600 1665 11:56:03 0 poll_s Ss+ /usr/bin/X -core :0 -auth /var/run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch
*j103 root 1600 -1 1600 1600 1767 11:56:13 0 wait Sl lightdm --session-child 16 19
*j104 lightdm 1791 -1 1791 1767 1791 11:56:15 0 sigsus Ss /bin/sh /usr/lib/lightdm/lightdm-greeter-session /usr/sbin/lightdm-gtk-greeter
*j105 lightdm 1791 -1 1791 1791 1797 11:56:18 0 poll_s Sl /usr/sbin/lightdm-gtk-greeter
*j106 root 1600 -1 1600 1600 1900 11:56:31 0 pipe_w S lightdm --session-child 12 19

instead i see messages as per a usual (noplymouth) reboot before lightdm awakens. i tried chvt (via ssh), seems strangely unable to do anything.

but i came back a couple hours later and the chvt *had* happened. so i tried again to "chvt 7". again the display immediately went into suspend mode. hitting keys doesn't reawaken it. "chvt 2" or other VTs followed by tapping keys also no effect (at least not within 10 or so minutes).

this has not happened before since installing trusty, so i'd guess it's not predictably repeatable.

my offhand guess fwiw is a latent bug in some display related driver. even back in natty i remember occasionally somehow falling into a situation where even chvt couldn't unfreeze the display.

ProblemType: Bug
DistroRelease: Ubuntu 14.04
Package: lightdm 1.9.3-0ubuntu1
ProcVersionSignature: Ubuntu 3.12.0-2.5-generic 3.12.0
Uname: Linux 3.12.0-2-generic i686
ApportVersion: 2.12.6-0ubuntu1
Architecture: i386
Date: Mon Nov 25 12:57:03 2013
LightdmConfig:
 [SeatDefaults]
 greeter-session=lightdm-gtk-greeter
 user-session=Lubuntu
LightdmGreeterLog:
 (lightdm-gtk-greeter:1797): GLib-CRITICAL **: g_shell_parse_argv: assertion 'command_line != NULL' failed

 ** (lightdm-gtk-greeter:1797): WARNING **: Failed to open sessions directory: Error opening directory '/usr/share/lightdm/sessions': No such file or directory

 ** (lightdm-gtk-greeter:1797): WARNING **: Failed to load user image: Failed to open file '/home/lucy/.face': No such file or directory
LightdmGreeterLogOld:
 (lightdm-gtk-greeter:1727): GLib-CRITICAL **: g_shell_parse_argv: assertion 'command_line != NULL' failed

 ** (lightdm-gtk-greeter:1727): WARNING **: Failed to open sessions directory: Error opening directory '/usr/share/lightdm/sessions': No such file or directory

 ** (lightdm-gtk-greeter:1727): WARNING **: Failed to load user image: Failed to open file '/home/lucy/.face': No such file or directory
MarkForUpload: True
SourcePackage: lightdm
UpgradeStatus: No upgrade log present (probably fresh install)

Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :

after another 20 or so minutes, tapping keys *does* awaken the display (i had left it at "chvt 2"). again i tried "chvt 7", and again i'm right back to a display in suspend mode that won't (soon) awaken.

Revision history for this message
gregrwm (gregrwm) wrote :

sure enough after rebooting again lightdm did come up properly this time.

Revision history for this message
Launchpad Janitor (janitor) wrote :

Status changed to 'Confirmed' because the bug affects multiple users.

Changed in lightdm (Ubuntu):
status: New → Confirmed
Revision history for this message
Alberto Salvia Novella (es20490446e) wrote :

Is estimated to have a moderate impact on a large portion of Ubuntu users.

Changed in lightdm (Ubuntu):
importance: Undecided → High
Revision history for this message
gregrwm (gregrwm) wrote :
Download full text (3.8 KiB)

i launched xscreensaver today, which i haven't done for awhile. within a half hour, the display locked up. likely it's the same bug. and likely that this bug has been around for many releases. back in natty i used to run xscreensaver more often, and used to see this bug alot. today i'm running saucy, as i do most of the time these days, trusty is just a tad too bleeding edge for daily work for me. but so anyway, today, the screen is still showing the antinspect screensaver, but the xscreensaver process is gone. completely. crashed apparently, along with several other X processes i had running.

the misbehaviour of the screen at this time is interesting. when i press ctl-alt-F7, i see the crashed antinspect xscreensaver. when i press ctl-alt-Fn, where n is {[12345689],10,11,12}, i see what i guess to be leftover from boot messages, perhaps what i should see (only!) on ctl-alt-F1. also seeing it in place of all the other VTs is bizarre.

at the same time that xscreensaver crashed, mplayer stopped playing (kfai) for about a minute, then resumed.

i can login via ssh and generate a process list, which does not contain xscreensaver nor the other X processes that crashed. attached. also i just happen to have had output redirected for most of those X apps. those outputs are also attached.

and here is a section from the "screen" session where i had launched various X apps which also died along with xscreensaver. (jackd didn't die, tho it did generate some complaints.) [1] was blueman-manager, [2] was linphone, [3] was pavucontrol, [4] was vncviewer, [5] was oclock, [7] was xscreensaver, [8] was firefox:

$ blueman-manager: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.

(pavucontrol:2453): Gdk-WARNING **: pavucontrol: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.0.

XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
      after 2925 requests (2685 known processed) with 0 events remaining.
[5] Exit 1 oclock -geometry 1019x975+1+-104 -fg green -transparent
JackEngine::XRun: client = alsa-jack.rawjackP.3652.0 was not finished, state = Triggered
JackEngine::XRun: client alsa-jack.rawjackP.3812.0 finished after current callback
JackAudioDriver::ProcessGraphAsyncMaster: Process error
JackEngine::XRun: client = alsa-jack.rawjackP.3652.0 was not finished, state = Triggered
JackEngine::XRun: client alsa-jack.rawjackP.3812.0 finished after current callback
JackAudioDriver::ProcessGraphAsyncMaster: Process error
[4] Exit 1 vncviewer -FullColor -PreferredEncoding=ZRLE -ZlibLevel=9 -AutoSelect=0 -MenuKey F7 $(vncviewer -h 2>&1|grep -q XDialog&&echo -XDialog) -geometry 1024x768--2--1 "$@" 2> /tmp/vncviewer$1:$(date +%F%a%T).err
[3] Exit 1 pavucontrol
[1] Exit 1 blueman-manager > /tmp/blueman-manager$(date +%F%a%T).err
[2] Exit 1 linphone 2> /tmp/linphone$(date +%F%a%T).err
JackEngine::XRun: client = alsa-jack.rawjackP.3652.0 was not finished, state = Triggered
JackAudioDriver::ProcessGraphAsyncMaster: Process error
[8]+ Exit 1 FLASH_ALSA_DEVICE=plug:SLAVE=rawjac...

Read more...

Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :
Revision history for this message
gregrwm (gregrwm) wrote :

alright, i don't *know* it's the same bug today. it certainly isn't a "lightdm" bug today. Xorg and display drivers are the common elements of suspect. if y'all would rather i filed today's episode as a separate bug just say so.

Revision history for this message
gregrwm (gregrwm) wrote :

i should say what leads me to presume they are the same bug is that the first thing i notice is that the ability to switch VTs is somehow broken.

Revision history for this message
Robert Ancell (robert-ancell) wrote :

Closing assumed fixed.

Changed in lightdm (Ubuntu):
status: Confirmed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.