Created attachment 160
workrave stderr output with --enable-debug

Workrave timers sit completely still in kubuntu jaunty. They worked fine in Intrepid.

launchpad bug report here: https://bugs.launchpad.net/ubuntu/+source/workrave/+bug/310551

I recompiled from source, using --enable-debug, and trying --enable-dbus (made no difference). I tried --enable-kde, but that requires kdelibs4-dev, which means I need to uninstall half my system...

The stderr output is attached.

The debug log indicates that you are idle all the time. Did you set the operation mode to 'suspended'? (look in the mode menu)

If not: What is the output of "xdpyinfo -queryExtensions | grep RECORD"

This should output something like: "RECORD  (opcode: 143, base error: 165)"

Nope, in normal mode.

$ xdpyinfo -queryExtensions | grep RECORD
    RECORD  (opcode: 135, base error: 144)

$ uname -a
Linux naught101-laptop 2.6.28-4-generic #5-Ubuntu SMP Fri Dec 26 22:48:51 UTC 2008 i686 GNU/Linux

I managed to reproduce the bug (using kubuntu jaunty installed in virtualbox).

It seems that Workrave does not get any events from the xrecord extension (except for 'start of data'). the xmacro package has the same problem: the xmacrorec2 program also does not get any data from xrecord.

I think the problem is caused by the recent xserver upgrade: my development PC, running Ubuntu Jaunty with out xserver upgrade, does not have this problem.

Changing to blocker, since this bug makes workrave completely useless (I'm guessing that this is a permanent change in the xserver?). Is there anything I can do to help figure out a fix?

The easiest workaround for now: uninstall libxtst-dev and reconfigure/recompile workrave. 

I did some googling, couldn't find any known issies with xrecord... needs further investigation.

hrm.. I removed libxtst-dev, and recompiled with:
./autogen-sh
./configuring
make
sudo checkinstall

I still have the same problem. Is there any particular options I should be configuring with?

Weird...

try this then:

after running configure. edit config.h.

comment out the following lines

/* Define if the RECORD extension is available */
#define HAVE_XRECORD

Rob, sorry for that last comment, I didn't realise, but libxtst hadn't uninstalled. I uninstalled it, and now workrave is working fine. 

Thanks so much, workrave is one of those vital pieces of software. It's easy to take for granted, until it goes missing...

I won't set the bug to resolved, since I'm not sure.

I finally found the problem: This X server no longer reports "core event" (like KeyPressed) through the record extension. Workrave now also intercepts device events from the XInputExtention (like DeviceKeyPressed).

This fix is not yet final. (I need to verify that Workrave still behaves normally on older X servers. I expect duplicate event that mess up the statistics...) 

The fix is in my git repository:

http://cgit.krandor.org/cgit.cgi?url=workrave/commit/&h=branch_v1_9&id=33f01a179da87e27dca3d57662872ef7077efb22

Fix tested and commited

*** Bug 830 has been marked as a duplicate of this bug. ***

I tried the autogen.sh from the git download, but get the error:

checking for DBUS... no
configure: Couldn't find d-bus.

I'm drawing a blank with google on this - any ideas?