Bug 1155 - blocks all user input except switch-to-terminal, but mouse moves - have to kill it with pkill
Status:
NEW
Component:
GUI
Version:
1.10
Hardware:
PC Linux
Importance:
P5 major
Target Milestone:
---
Assignee:
Rob Caelers
URL:
Depends on:
Blocks:
Reported:
Feb 28 2014 09:03:24 UTC
by:
Arne Babenhauserheide
Modified:
Oct 25 2017 14:17:55 UTC
CC List:
18**@gm**.com
Rance
| Who | When | What | Removed | Added |
|---|---|---|---|---|
| Arne Babenhauserheide | Jun 5 2014 05:20:19 UTC | platform | All | PC |
| op_sys | All | Linux | ||
| severity | normal | major | ||
| Rance | Jun 23 2015 08:03:29 UTC | cc | Rance | |
| 18**@gm**.com | Oct 25 2017 14:17:55 UTC | cc | 18**@gm**.com |
Description
Arne Babenhauserheide Feb 28 2014 09:03:24 UTC
On my Gentoo KDE GNU Linux system workrave blocked all user-interaction except for moving the mouse (but not clicking) and leaving X11 with ctrl-alt-F1 (or other F-keys). To get my KDE session responsive again, I have to switch to the terminal and kill workrave with pkill -3 workrave. block-mode is set to None in the config.
Comment 1
Arne Babenhauserheide Feb 28 2014 09:08:36 UTC
This seems to only happen in normal breaks - at least the first micro-break works normally.
Comment 2
Arne Babenhauserheide Jun 5 2014 05:18:33 UTC
Yesterday I tried workrave again and when I ignored the microbreak for more than 5-6 minutes it blocked all user-input. This was still more than 10 minutes before the next microbreak. I can now trigger it intentionally by waiting out for the maximum number of microbreak warnings and then hitting skip. Then it counts down to the next micro-break and just before the micro-break timer hits 0 it stops the screen from updating. My user-input actually still is accepted (I can simply keep typing), but I no longer see any updates on the screen. To check this, I set the wait-time to the microbreak to 3s and the “postpone”-setting to 4s. The microbreak-length is still 30s.
Comment 3
Arne Babenhauserheide Jun 5 2014 05:20:19 UTC
Set the importance to major, because it makes it impossible for me to work with workrave, and because I can now trigger it intentionally, so it could be possible to fix it.
Comment 4
Arne Babenhauserheide Jul 14 2014 05:02:35 UTC
I now found a workaround: When I disable the checkbox “maximum number of […]”, then my screen does not freeze. I now did that. It makes workrave a bit more annoying and disables a feature, but it enables me to use it again.
Comment 5
Arne Babenhauserheide Dec 23 2014 08:44:20 UTC
I’m not sure what happened, but the problem appeared again in August. I still only have 1.10.1, though - need to get the Gentoo ebuild updated.
Comment 6
Arne Babenhauserheide Dec 23 2014 11:14:34 UTC
There’s now a new ebuild for Gentoo[1] (I created it), but I cannot test 1.10.6 yet, because there’s no release on github. I’d like to test whether the blocking changes I saw in the NEWS file fix my issues. As soon as you release 1.10.6, that test will only require renaming a file. [1]: https://bugs.gentoo.org/show_bug.cgi?id=533376
Comment 7
Rob Caelers Dec 23 2014 15:11:20 UTC
Hi. 1.10.6 hasn't been released yet. The major difference between 1.10.5 and 1.10.6 will be applets for xfce, cinnamon and mate. I have never been able to reproduce this issue (on Ubuntu). If blocking mode if set to 'Input only' or 'Input and screen' Workrave should block keyboard and mouse clicks, but only during breaks. What is your blocking mode settings? If blocking mode is set to 'input and screen', Workrave replaces your screen with only the background. All windows (except the break window) should disappear. Is this what you are seeing? Or do all windows remain visible, but are just no updated anymore? I will try to install Gentoo and see whether I can reproduce this problem.
Comment 8
Arne Babenhauserheide Jan 15 2015 08:33:55 UTC
Status update: I now got workrave 1.10.6 into Gentoo. My blocking mode is to not block anything (I don’t need it to block, I just need it to nag until I take a break ☺).
Comment 9
Arne Babenhauserheide Jan 15 2015 08:36:03 UTC
(In reply to ro**@kr**.org from comment #7) > If blocking mode is set to 'input and screen', Workrave replaces your screen > with only the background. All windows (except the break window) should > disappear. Is this what you are seeing? Or do all windows remain visible, > but are just no updated anymore? For me all windows remain visible, but are just no updated anymore? It also seems like the computer still takes input (after killing workrave I saw changes), it just does not update the screen anymore.
Comment 10
Rance Jun 23 2015 08:03:29 UTC
I have been having this problem as well for a long time. I am using the same Gentoo with AwesomeWM window manager. I have not found that messing with any of the options stops it from hanging X I can remotely login to the computer and kill the process. I have run it through GDB last time it locked up and this the the BT output, although it does not really look too helpful. Let me know what other things I can do to figure out where it might be stopping. #0 0x00007ffff3bea6cc in __lll_lock_wait () from /lib64/libpthread.so.0 #1 0x00007ffff3be5ed1 in pthread_mutex_lock () from /lib64/libpthread.so.0 #2 0x00007ffff729fd57 in XSetErrorHandler () from /usr/lib64/libX11.so.6 #3 0x00007ffff576e12c in ?? () from /usr/lib64/libgdk-3.so.0 #4 0x00007ffff576510e in gdk_x11_display_error_trap_push () from /usr/lib64/libgdk-3.so.0 #5 0x00007ffff5760366 in ?? () from /usr/lib64/libgdk-3.so.0 #6 0x00007ffff574476a in gdk_device_get_window_at_position_double () from /usr/lib64/libgdk-3.so.0 #7 0x00007ffff5744835 in gdk_device_get_window_at_position () from /usr/lib64/libgdk-3.so.0 #8 0x00007ffff5c53616 in gtk_tooltip_trigger_tooltip_query () from /usr/lib64/libgtk-3.so.0 #9 0x00007ffff4664d35 in g_slist_foreach () from /usr/lib64/libglib-2.0.so.0 #10 0x00007ffff5c89fbf in ?? () from /usr/lib64/libgtk-3.so.0 #11 0x00007ffff5741320 in ?? () from /usr/lib64/libgdk-3.so.0 #12 0x00007ffff464b380 in g_main_context_dispatch () from /usr/lib64/libglib-2.0.so.0 #13 0x00007ffff464b673 in ?? () from /usr/lib64/libglib-2.0.so.0 #14 0x00007ffff464b905 in g_main_loop_run () from /usr/lib64/libglib-2.0.so.0 #15 0x00007ffff5b7afb4 in gtk_main () from /usr/lib64/libgtk-3.so.0 #16 0x00000000004459f1 in GUI::main() () #17 0x0000000000472714 in run () #18 0x00007ffff386fa5c in __libc_start_main () from /lib64/libc.so.6 #19 0x0000000000447ee5 in _start () Thanks
Comment 11
Rance Jan 27 2016 08:09:38 UTC
FYI I have found now with my current version that I no longer have this issue.
* app-misc/workrave
Latest version available: 1.10.8
Latest version installed: 1.10.8
* x11-wm/awesome
Latest version available: 3.5.6-r2
Latest version installed: 3.4.15
* x11-base/xorg-server
Latest version available: 1.17.4
Latest version installed: 1.17.4
Hope this helps at all.Comment 12
Arne Babenhauserheide Jul 11 2016 16:22:43 UTC
(In reply to Rance from comment #11)
> FYI I have found now with my current version that I no longer have this
> issue.
>
> * app-misc/workrave
> Latest version available: 1.10.8
> Latest version installed: 1.10.8
>
> * x11-wm/awesome
> Latest version available: 3.5.6-r2
> Latest version installed: 3.4.15
>
> * x11-base/xorg-server
> Latest version available: 1.17.4
> Latest version installed: 1.17.4
>
>
> Hope this helps at all.
I also no longer have the issue (I just came here to report that when I tested it again last week, it was gone).,
* app-misc/workrave
versions: 1.10.6-r1 1.10.6-r2 1.10.8
installed: 1.10.8
* x11-base/xorg-server
versions: 1.12.4-r5 1.12.4-r7 1.15.2-r2 1.15.2-r4 1.16.4 1.16.4-r5 1.17.4 1.18.0 1.18.1 1.18.2 1.18.3
installed: 1.17.4
* kde-base/kwin
versions: 4.11.22
installed: 4.11.22
* kde-apps/kde-meta
versions: 4.14.3 4.14.3-r1 15.08.3 15.12.3 16.04.1
installed: 4.14.3-r1
* kde-base/plasma-workspace
versions: 4.11.22
installed: 4.11.22