Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
[...]
Updating the machine hasn't been easy and hasn't fixed the problem either. Evolution dep breakage required me to update wildcard-groups of packages or else I couldn't get xorg* and other updates. Applying the xorg* updates required a reboot as a simple logout/login resulted in severe screen corruption.
[...]
During the boot process (this is ATI Radeon graphics hardware), I often lose the display and get a black screen with my monitor losing signal. This has never happened with F14 and older, and only rarely with F15. A workaround is to disable rhgb.
Michael,
I've been having recent issues with a Radeon HD 2600 card. Could you review your messages logs for entries containing : * page flip * unable to schedule IB * wait for fifo failed ... associated with drm or radeon.
You might be having the same problem as myself....
Pat.
On 08/28/2011 04:05 PM, Michael Schwendt wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
[...]
Updating the machine hasn't been easy and hasn't fixed the problem either. Evolution dep breakage required me to update wildcard-groups of packages or else I couldn't get xorg* and other updates. Applying the xorg* updates required a reboot as a simple logout/login resulted in severe screen corruption.
[...]
During the boot process (this is ATI Radeon graphics hardware), I often lose the display and get a black screen with my monitor losing signal. This has never happened with F14 and older, and only rarely with F15. A workaround is to disable rhgb.
On Sun, 28 Aug 2011 16:12:27 -0400, PD (Patrick) wrote:
I've been having recent issues with a Radeon HD 2600 card.
This is with Radeon HD 3450 (aka RV620).
Could you review your messages logs for entries containing :
- page flip
- unable to schedule IB
- wait for fifo failed
... associated with drm or radeon.
You might be having the same problem as myself....
Found something else with a series of "grep" searches. GConf2 crashes:
# grep segfault messages Aug 28 21:22:46 localhost kernel: [ 6534.040823] at-spi-registry[21145]: segfault at 18 ip 0000003d5a219951 sp 00007fff42a99160 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:24:08 localhost kernel: [ 40.220700] at-spi-registry[1153]: segfault at 18 ip 0000003d5a219951 sp 00007fffb76da8e0 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:24:36 localhost kernel: [ 68.686891] at-spi-registry[1737]: segfault at 18 ip 0000003d5a219951 sp 00007fff96f150c0 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:30:57 localhost kernel: [ 38.605513] at-spi-registry[1206]: segfault at 18 ip 0000003d5a219951 sp 00007fffb870eb30 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:31:33 localhost kernel: [ 74.177443] at-spi-registry[1718]: segfault at 18 ip 0000003d5a219951 sp 00007fff45001660 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:35:56 localhost kernel: [ 337.510879] at-spi-registry[2751]: segfault at 18 ip 0000003d5a219951 sp 00007fff56a32a40 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:38:08 localhost kernel: [ 39.329801] at-spi-registry[1149]: segfault at 18 ip 0000003d5a219951 sp 00007fffbcc77e00 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:38:43 localhost kernel: [ 75.030405] at-spi-registry[1809]: segfault at 18 ip 0000003d5a219951 sp 00007fff72a20900 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:39:49 localhost kernel: [ 141.127202] at-spi-registry[2381]: segfault at 18 ip 0000003d5a219951 sp 00007fff25eec820 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:40:16 localhost kernel: [ 167.889173] at-spi-registry[2867]: segfault at 18 ip 0000003d5a219951 sp 00007fff00a43890 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:40:28 localhost kernel: [ 179.922760] at-spi-registry[3280]: segfault at 18 ip 0000003d5a219951 sp 00007fff60b30410 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000]
On Sun, Aug 28, 2011 at 1:21 PM, Michael Schwendt mschwendt@gmail.com wrote:
On Sun, 28 Aug 2011 16:12:27 -0400, PD (Patrick) wrote:
I've been having recent issues with a Radeon HD 2600 card.
This is with Radeon HD 3450 (aka RV620).
Could you review your messages logs for entries containing :
- page flip
- unable to schedule IB
- wait for fifo failed
... associated with drm or radeon.
You might be having the same problem as myself....
Found something else with a series of "grep" searches. GConf2 crashes:
# grep segfault messages Aug 28 21:22:46 localhost kernel: [ 6534.040823] at-spi-registry[21145]: segfault at 18 ip 0000003d5a219951 sp 00007fff42a99160 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:24:08 localhost kernel: [ 40.220700] at-spi-registry[1153]: segfault at 18 ip 0000003d5a219951 sp 00007fffb76da8e0 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:24:36 localhost kernel: [ 68.686891] at-spi-registry[1737]: segfault at 18 ip 0000003d5a219951 sp 00007fff96f150c0 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:30:57 localhost kernel: [ 38.605513] at-spi-registry[1206]: segfault at 18 ip 0000003d5a219951 sp 00007fffb870eb30 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:31:33 localhost kernel: [ 74.177443] at-spi-registry[1718]: segfault at 18 ip 0000003d5a219951 sp 00007fff45001660 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:35:56 localhost kernel: [ 337.510879] at-spi-registry[2751]: segfault at 18 ip 0000003d5a219951 sp 00007fff56a32a40 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:38:08 localhost kernel: [ 39.329801] at-spi-registry[1149]: segfault at 18 ip 0000003d5a219951 sp 00007fffbcc77e00 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:38:43 localhost kernel: [ 75.030405] at-spi-registry[1809]: segfault at 18 ip 0000003d5a219951 sp 00007fff72a20900 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:39:49 localhost kernel: [ 141.127202] at-spi-registry[2381]: segfault at 18 ip 0000003d5a219951 sp 00007fff25eec820 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:40:16 localhost kernel: [ 167.889173] at-spi-registry[2867]: segfault at 18 ip 0000003d5a219951 sp 00007fff00a43890 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] Aug 28 21:40:28 localhost kernel: [ 179.922760] at-spi-registry[3280]: segfault at 18 ip 0000003d5a219951 sp 00007fff60b30410 error 4 in libgconf-2.so.4.1.5[3d5a200000+2f000] --
Don't think this is it, as I see these as well, but I have no problems logging into gnome:
Aug 27 09:46:01 tlondon kernel: [ 50.870783] at-spi-registry[1368]: segfault at 18 ip 000000306fa19951 sp 00007fff48224400 error 4 in libgconf-2.so.4.1.5[306fa00000+2f000] Aug 27 10:15:36 tlondon kernel: [ 50.433420] at-spi-registry[1365]: segfault at 18 ip 000000306fa19951 sp 00007fff4ab16df0 error 4 in libgconf-2.so.4.1.5[306fa00000+2f000] Aug 28 09:01:24 tlondon kernel: [ 49.896743] at-spi-registry[1356]: segfault at 18 ip 000000306fa19951 sp 00007fff934ae540 error 4 in libgconf-2.so.4.1.5[306fa00000+2f000] Aug 28 09:15:37 tlondon kernel: [ 902.539762] at-spi-registry[2746]: segfault at 18 ip 00007fda267929d1 sp 00007fffec41d420 error 4 in libgconf-2.so.4.1.5[7fda26779000+2f000] Aug 28 12:25:14 tlondon kernel: [ 51.556350] at-spi-registry[1361]: segfault at 18 ip 00007fc7deeb49d1 sp 00007fffeb8220a0 error 4 in libgconf-2.so.4.1.5[7fc7dee9b000+2f000]
tom
On 08/29/2011 01:35 AM, Michael Schwendt wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
I had hit this issue during F15 upgrade and looking at the .xsession-erros, looks like you will have to "yum remove gnome-screensaver"
~Sudhir.
[...]
Updating the machine hasn't been easy and hasn't fixed the problem either. Evolution dep breakage required me to update wildcard-groups of packages or else I couldn't get xorg* and other updates. Applying the xorg* updates required a reboot as a simple logout/login resulted in severe screen corruption.
[...]
During the boot process (this is ATI Radeon graphics hardware), I often lose the display and get a black screen with my monitor losing signal. This has never happened with F14 and older, and only rarely with F15. A workaround is to disable rhgb.
On Mon, 29 Aug 2011 10:44:54 +0530, SD (Sudhir) wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
I had hit this issue during F15 upgrade and looking at the .xsession-erros, looks like you will have to "yum remove gnome-screensaver"
Didn't fix it. What would be the rationale?
Several other programs mention the same problems in .xsession-errors After "rpm -e gnome-screensaver" I've had five consecutive "Oh no" login attempts, and the 6th worked. Race condition with something crashing or failing to initialise?
Dne 29.8.2011 15:32, Michael Schwendt napsal(a):
On Mon, 29 Aug 2011 10:44:54 +0530, SD (Sudhir) wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
I had hit this issue during F15 upgrade and looking at the .xsession-erros, looks like you will have to "yum remove gnome-screensaver"
Didn't fix it. What would be the rationale?
Several other programs mention the same problems in .xsession-errors After "rpm -e gnome-screensaver" I've had five consecutive "Oh no" login attempts, and the 6th worked. Race condition with something crashing or failing to initialise?
Well I'm no where near being gnome expert - but this is the way I'm doing my debugging of frequent gnome-shell issue/problems/bugs.
I change my ~/.xinitrc to only exec:
exec gnome-terminal
(eventually as a last instance exec xterm) (and I'm also using startx instead of default login display manager)
Within this terminal I try to play with gnome - i.e. start gnome-shell and check ongoing messages and started programs and related bugs.
Zdenek
On 08/29/2011 07:02 PM, Michael Schwendt wrote:
On Mon, 29 Aug 2011 10:44:54 +0530, SD (Sudhir) wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
I had hit this issue during F15 upgrade and looking at the .xsession-erros, looks like you will have to "yum remove gnome-screensaver"
Didn't fix it. What would be the rationale?
Several other programs mention the same problems in .xsession-errors
I'd try to remove those packages too. What I suspect is you have some kinda theme running?
After "rpm -e gnome-screensaver" I've had five consecutive "Oh no" login
Any reason why you are not using yum?
attempts, and the 6th worked. Race condition with something crashing or failing to initialise?
On Mon, 29 Aug 2011 20:11:37 +0530, SD (Sudhir) wrote:
I had hit this issue during F15 upgrade and looking at the .xsession-erros, looks like you will have to "yum remove gnome-screensaver"
Didn't fix it. What would be the rationale?
Several other programs mention the same problems in .xsession-errors
I'd try to remove those packages too.
If I removed everything that logs some sort of warning in .xsession-errors, I could remove the entire DE.
What I suspect is you have some kinda theme running?
Default GNOME Shell.
After "rpm -e gnome-screensaver" I've had five consecutive "Oh no" login
Any reason why you are not using yum?
It doesn't matter for a package with no dependencies.
On Sun, 2011-08-28 at 22:05 +0200, Michael Schwendt wrote:
Never have seen this with F15. Only have read about it.
Now it's hitting me badly everytime I boot F16 Alpha. Sometimes the second login attempt succeeds, sometimes it doesn't and needs a third attempt.
gnome-shell-3.1.4-1.fc16.x86_64
Where do I search for meaningful error messages? Attached .xession-errors is flooded.
[...]
Updating the machine hasn't been easy and hasn't fixed the problem either. Evolution dep breakage required me to update wildcard-groups of packages or else I couldn't get xorg* and other updates. Applying the xorg* updates required a reboot as a simple logout/login resulted in severe screen corruption.
[...]
During the boot process (this is ATI Radeon graphics hardware), I often lose the display and get a black screen with my monitor losing signal. This has never happened with F14 and older, and only rarely with F15. A workaround is to disable rhgb.
GNOME devs refer to the screen as the 'fail whale'. It shows up if any key Shell component crashes in a loop. The key message to look for is "respawning too quickly", thus:
(gnome-settings-daemon:1639): GnomeDesktop-WARNING **: could not get output property for DVI-0, rc: 15
(gnome-settings-daemon:1639): GLib-CRITICAL **: g_variant_new_string: assertion `string != NULL' failed
(gnome-settings-daemon:1639): color-plugin-WARNING **: failed to enumerate directory /home/misc16/.color/icc: No such file or directory g_dbus_connection_real_closed: Remote peer vanished with error: Underlying GIOStream returned 0 bytes on an async read (g-io-error-quark, 0). Exiting. gnome-session[1328]: WARNING: App 'gnome-settings-daemon.desktop' respawning too quickly
so that's your problem there, I think. alt-f4 should get rid of the fail whale, and abrt-gui should be able to report the g-s-d crash.
You're not running glibc 2.14.90-6 , are you? It's known to be broken. I ran it for a bit today and found g-s-d would crash.
On Mon, 29 Aug 2011 20:11:04 -0700, AW (Adam) wrote:
GNOME devs refer to the screen as the 'fail whale'. It shows up if any key Shell component crashes in a loop. The key message to look for is "respawning too quickly", thus:
(gnome-settings-daemon:1639): GnomeDesktop-WARNING **: could not get output property for DVI-0, rc: 15
(gnome-settings-daemon:1639): GLib-CRITICAL **: g_variant_new_string: assertion `string != NULL' failed
(gnome-settings-daemon:1639): color-plugin-WARNING **: failed to enumerate directory /home/misc16/.color/icc: No such file or directory g_dbus_connection_real_closed: Remote peer vanished with error: Underlying GIOStream returned 0 bytes on an async read (g-io-error-quark, 0). Exiting. gnome-session[1328]: WARNING: App 'gnome-settings-daemon.desktop' respawning too quickly
so that's your problem there, I think. alt-f4 should get rid of the fail whale, and abrt-gui should be able to report the g-s-d crash.
Hey! This has changed the world here. With Alt-F4 I continued, opened a terminal and ran /usr/libexec/gnome-settings-daemon &> g-s-d.LOG Initially, it printed the same warnings as in xsession-errors and then terminated without crashing. Therefore ABRT didn't intercept anything. Afterwards I've run into other issues (one being a reproducible crash in gnome-shell I've reported) and small display corruption, which made me reboot, and now I cannot reproduce the g-s-d problem at login time. Not with my "test1" account either. I've restored a backup of my "misc16" account files, and still cannot reproduce it anymore.
You're not running glibc 2.14.90-6 , are you? It's known to be broken. I ran it for a bit today and found g-s-d would crash.
No, -4 and never had -6 installed.
On Mon, 2011-08-29 at 20:11 -0700, Adam Williamson wrote:
GNOME devs refer to the screen as the 'fail whale'.
The difference is, Twitter has millions (billions?) of users, most of whom are non-technical and explicitly don't want to know what went wrong with *Twitter's* servers. Most desktop Linux (not Android, etc.) users I know are more technical and would at least like some information about what went wrong with *their* own computer.
Jon.
-
Adam Williamson -
Jon Masters -
Michael Schwendt -
Patrick Dubois -
Sudhir D -
Tom London -
Zdenek Kabelac