window selection problems

[Rkimber]

I have Dapper with latest updates on dual opteron and 2.6.15-23-amd64-generic kernel.

The option in the system/preferences/windows for selecting windows when mouse moves over them doesn't seem to work.

Also, when this preference is not enabled (i.e. when I assume I'm using click-to-raise) clicking in a window sometimes has no effect. Usually in this situation it takes a click on the title bar after having clicked elsewhere to bring the window to the top.

[Sebastien Bacher (seb128)]

Thanks for your bug. What do you mean by "doesn't seem to work", does it actually work then? Could you describe some steps to get the issue? You should not need to click if you use the mouseover option ...

[Rkimber (rkimber)]

What I meant by 'doesn't seem to' is that it doesn't work, but I was allowing the possibility that I was doing something wrong. I think the rest should be clear. There are two problems:

1) changing the preference option to having the window made current when the mouse moves over it doesn't have any effect. The steps you take are mentioned above.

2) when the option mentioned in (1) is *not* supposed to be in effect - i.e. when you are using click to raise the window, clicking in a window sometimes has no effect. By saying 'sometimes' I mean that sometimes it works and sometimes it doesn't and I can't find any clear way of reproducing it consistently - other than just using the system. It happens frequently enough to be distracting (every few window changes, where 'few' means somewhere between, say, 4 and 10 though I don't have a record, and I'm currently using 5.10 because it's much more stable in other respects).

[Lothy (christina-lucy)]

I have a problem that I believe is the same as this one. I have posted about it at ubuntuforums.org, you can read here for details - http://ubuntuforums.org/showthread.php?t=196336

[Elijah (newren)]

This sounds like bug 57344; it was a race condition most people were not able to trigger with metacity versions prior to 2.15.34, but which became easier to trigger with that version. Differences in hardward definitely affect the ability to reproduce race conditions such as these, and you may have just been one of the unlucky ones. There's a patch in http://bugzilla.gnome.org/show_bug.cgi?id=354422 that should fix this for you, if I'm right about my guess.

[Sebastien Bacher (seb128)]

Marking as duplicate of bug #57344 as pointed by Elijah. A package with the mentionned patch has been upload, let we know if it fixes your issue