Description of problem: When connected to a Windows guest (with mouse in server mode (no vdagent) from a windows client mouse pointer tends to disappear as parts of remote-viewer get moved outside of visible area. Version-Release number of selected component (if applicable): mingw-virt-viewer-0.5.3-16 How reproducible: Always Steps to Reproduce: 1. Connect to a windows 7 guest without vdagent 2. Move remote viewer towards left edge of the screen, so that just less than a half of it is visible 3. Try doing something in the guest using mouse Actual results: Cursor doesn't move the way it should, and if moved towards right edge of guest it disappears.
I really can't manage to reproduce with current mingw-virt-viewer-0.5.3-16.el6
Can you record a screencast? Do you have dual-head or any setup different from simple traditional desktop & guest? thanks
Created attachment 644085 [details] And for my final trick I will make the cursor disappear There shouldn't be anything special about the setup. Single monitor, vdagent is disabled.
(In reply to comment #3) > Created attachment 644085 [details] > And for my final trick I will make the cursor disappear > > There shouldn't be anything special about the setup. Single monitor, vdagent > is disabled. how do you move the window? it looks like the pointer is still grabbed by remote-viewer but you managed to escape it the second time(like alt+click+move on unix), so you end up with 2 cursors being drawn. pretty weird, I don't know what's the trick..
> how do you move the window? it looks like the pointer is still grabbed by > remote-viewer but you managed to escape it the second time(like > alt+click+move on unix), so you end up with 2 cursors being drawn. pretty > weird, I don't know what's the trick.. The trick is, that the regular client cursor isn't captured in the screencast. Cursor is released when moving the window (you can tell by the changes in titlebar).
hey, can you try againt with latest build? mingw-virt-viewer-0.5.3-17 It could eventually be fixed with bug 872640. thanks for checking
Well, bad news. The bug is still there.
(In reply to comment #5) > > how do you move the window? > The trick is, that the regular client cursor isn't captured in the > screencast. Cursor is released when moving the window (you can tell by the > changes in titlebar). how do you move the window without dragging the title bar? please explain in more details what you do, I still can't reproduce.
I could reproduced, strangely inside a VM win7 atm. However the bug description is a bit misleading. It is not just the cursor that is not drawn, it is the whole drawing area. It looks like a weird Gtk or Cairo bug to me, investigating now.
It's also only left-sided bug..
more weirdish stuff, if gtk_widget_set_has_window() is FALSE, then the bug don't happen. Looks to me like a Gdk bug now
so setting the widget double buffered is enough to fix it, but I don't know what will be the side-effect. I will enable it in the windows build for now, and open a bug in gtk+
https://bugzilla.gnome.org/show_bug.cgi?id=688962
please add acks!
patch added, moving bug to modified
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHEA-2013-0889.html