Screensaver disables visual effects

[Luis Fernando Teixeira]

Steps to reproduce the error:

1) working normally with visual effects set to "Extra"

2) activate screensaver via lock screen gnome panel applet

3) unlock screen, entering password

4) workspace appears, but visual effects are set to "None"

The visual effects can be set to "Extra" again, no problems, just boring...

I am using ubuntu 8.04 Hardy Heron (desktop), updated to 2008/06/12. I use the nvidia drivers (nvidia-glx-new 169.12+2.6.24.13-18.41).

[Pedro Villavicencio (pedro)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. You reported this bug a while ago and there hasn't been any activity in it recently. We were wondering is this still an issue for you? Could you try to reproduce the same with Ubuntu 8.10 or 9.04? Thanks in advance.

[Pedro Villavicencio (pedro)]

We are closing this bug report because it lacks the information we need to investigate the problem, as described in the previous comments. Please reopen it if you can give us the missing information, and don't hesitate to submit bug reports in the future. To reopen the bug report you can click on the current status, under the Status column, and change the Status back to New. Thanks again!.

[wfaulk (wfaulk)]

This bug affects me, though it's not 100% reproducible. It happens some times and not other times, with no apparent connection to length of screensaver activation, whether power saving kicks in, or anything else I can think of. Sometimes it happens after a few minutes, sometimes days of screensaver activation won't cause it. But I've never seen it happen without the screensaver being active.

If this bug can be reopened, I'd be happy to provide whatever input I can.

[wfaulk (wfaulk)]

I reopened the bug. I'd also like to add that when this happens, my workspace switcher changes geometry. I usually have it set up to have 3 rows and 3 columns. When the visual effects drop back to "none", it switches to 1 row and 9 columns. This doesn't happen when I change the visual effects myself. Also, if I change the geometry back before I change the effects setting, it flattens it again.

It might also be important to note that the Appearances GUI doesn't show any of the three visual effects radio buttons as active when the "extra" effects are disabled.

[wfaulk (wfaulk)]

I've noticed that when it happens now, the "None" button is selected. I don't know if something changed or if my earlier assertion about the radio buttons was just incorrect.

[wfaulk (wfaulk)]

Oh, and I just stumbled across this data from dmesg:

[432685.442663] plasma[12126]: segfault at 8 ip 00007f9da5eb0583 sp 00007fff78d30400 error 4 in libGLcore.so.185.18.36[7f9da53be000+dda000]

and

[486068.000875] compiz.real[8091]: segfault at 10 ip 00007f35f8298f07 sp 00007fffa4f817c8 error 4 in libGLcore.so.185.18.36[7f35f77a2000+dda000]

plasma is the screensaver module I'm running, and it seems to segfault frequently.

[wfaulk (wfaulk)]

Unsurprisingly, it looks like the nvidia drivers are at fault:

% dpkg-query -S /usr/lib/libGLcore.so.1
nvidia-glx-185: /usr/lib/libGLcore.so.1

[Chris Coulson (chrisccoulson)]

It's not a screensaver bug if compiz is crashing. And the information you've provided shows that it is probably crashing somewhere in the NVIDIA driver, although you've not provided a full backtrace. I'll reassign it for now though. There's generally not much anybody can do about crashes in the proprietary NVIDIA driver anyway though

[wfaulk (wfaulk)]

Well, you could argue that a compiz crash shouldn't replace compiz with metacity, unless it's crashing immediately. Also, the geometry of the workspace switcher shouldn't change.

[Chris Coulson (chrisccoulson)]

The crash handling in compiz calls the fallback window manager (currently metacity by default), to stop it from crashing repeatedly. I think you can change that locally in your preferences if you prefer too though. The geometry of the workspace switcher is a window manager specific preference, so if you launch a new window manager, then it will use it's own preferences. If you want Compiz and Metacity to have the same number of workspaces, you need to change their preferences.

[wfaulk (wfaulk)]

Re: fallback manager: fair enough.

Re: workspace switcher: If I switch back and forth manually, the workspace switcher geometry remains the same (3x3). Only when compiz crashes do I get the 9x1 geometry. Perhaps it is calling metacity as a fallback manager with a different configuration somehow?

Actually, if I manually send a SIGSEGV to compiz.real, it doesn't change the geometry. Weird. I'm out of ideas.

nVidia recently released a gold version of their v190 drivers, which I just upgraded to. Hopefully I won't see this problem again.

[Chris Coulson (chrisccoulson)]

Yeah, hopefully. Unfortunately, there's not much else I can do with crashes in nvidia drivers, other than encourage you to bug nvidia about it ;)

[wfaulk (wfaulk)]

I'd been running without any problems since Dec10 with the v190 drivers at the nVidia repo at http://ppa.launchpad.net/nvidia-vdpau/ppa/ubuntu. Today I came into work to find that the v190 drivers were available from an Ubuntu repository. I had them running for about two hours before I encountered the same problem again.

I would have figured that Ubuntu was just repackaging the drivers from nVidia, but clearly something different is happening.

[bugbot (bugbot)]

This bug report was filed against an old version of Ubuntu.
Can you confirm whether this is still an issue in natty?

If you don't mind, it would be very helpful if you could update the bug
report in launchpad to 'Fix Released' if it is no longer an issue for
you, or if it is still occurring under natty, please tag the bug 'natty'
so it's easier for us to track.

[bugbot (bugbot)]

We're closing this bug since it is has been some time with no response from the original reporter. However, if the issue still exists please feel free to reopen with the requested information. Also, if you could, please test against the latest development version of Ubuntu, since this confirms the bug is one we may be able to pass upstream for help.