Bug 1018182

Summary:	primary monitor is switched if some screen gets bigger then current primary screen
Product:	Red Hat Enterprise Virtualization Manager	Reporter:	David Jaša <djasa>
Component:	mingw-virt-viewer	Assignee:	Jonathon Jongsma <jjongsma>
Status:	CLOSED ERRATA	QA Contact:	Desktop QE <desktop-qa-list>
Severity:	high	Docs Contact:
Priority:	unspecified
Version:	3.3.0	CC:	cfergeau, dblechte, jjongsma, juwu, marcandre.lureau, mkrcmari, rbalakri, yeylon
Target Milestone:	---
Target Release:	3.5.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	mingw-virt-viewer-0.6.0-1	Doc Type:	Bug Fix
Doc Text:	Display configuration sometimes used outdated information about the position of the remote-viewer windows in order to align and configure the guest displays. Occasionally, this caused the guest displays to became unexpectedly swapped when a window is resized. With this update, remote-viewer will always use the current window locations to align displays, rather than using a possibly outdated cached location information.	Story Points:	---
Clone Of:	1018180	Environment:
Last Closed:	2015-02-11 17:43:23 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1018180
Bug Blocks:	1009648, 1142923, 1156165

Description David Jaša 2013-10-11 11:45:26 UTC

This bug affects windows clients as well as linux clients. Any discussion relevant to both bugs should take place at the parent bug.


+++ This bug was initially created as a clone of Bug #1018180 +++

Description of problem:
when user resizes monitors in a way that current primary monitors gets smaller than some other monitor, the now-biggest screen becomes primary. This is quite confusing for the users and it should not happen.
The issue occurs just with new clients, I didn't manage to reproduce with old (3.2) client/new vdagent combo.

Version-Release number of selected component (if applicable):
RHEL 6.5: spice-gtk-0.20-9.el6.x86_64
Windows 7 32b: mingw-virt-viewer-0.5.6-6.el6_64

How reproducible:
with 3.3 vdagent: roughly 70-100 % of cases
with 3.2 vdagent: occurs in one of multiple tries

Steps to Reproduce:
1. connect to multiple monitor guest (windowed)
2. resize primary monitor (with taskbar) to be smaller than some other monitor (or vice versa - resize non-primary monitor to be larger than primary)
3.

Actual results:
the now-biggest monitor becomes primary

Expected results:
primary monitor is not changed at all

Additional info:
The only case when primary monitor is changed is when current primary monitor is disabled - guest should however handle this condition entirely itself

Comment 1 Marc-Andre Lureau 2013-11-07 13:39:51 UTC

needinfo, discussion in dup bug.

Comment 2 David Blechter 2013-12-08 20:56:46 UTC

pushing to 3.4 as no blocker flag proposed and we are at RC phase and the issue is under investigation

Comment 3 Marc-Andre Lureau 2014-03-04 18:32:40 UTC

parent bug is in POST, moving here too

Comment 4 Marc-Andre Lureau 2014-03-04 18:32:51 UTC

parent bug is in POST, moving here too

Comment 5 Marc-Andre Lureau 2014-04-17 10:57:39 UTC

moving to Jonathon, who fixed the bug

Comment 6 David Blechter 2014-04-17 19:46:40 UTC

We'll re-base in in the next release, moving to 3.5

Comment 9 Julie 2015-01-27 07:22:44 UTC

hi Marc-Andre,
 Would you be able to explain a bit more about "Fix: Always use the current window location to align and configure displays". What are we recommending users to do to avoid the screen swap?

Cheers,
Julie

Comment 10 Marc-Andre Lureau 2015-01-27 10:22:25 UTC

(In reply to Julie from comment #9)
> hi Marc-Andre,
>  Would you be able to explain a bit more about "Fix: Always use the current
> window location to align and configure displays". What are we recommending
> users to do to avoid the screen swap?
> 
> Cheers,
> Julie

I used Jonathon comments from bug 1018180, better ask him.

Comment 11 Jonathon Jongsma 2015-01-27 15:12:50 UTC

Sorry for the confusing wording. The user does not need to do anything. My comment simply means that virt-viewer itself will always use the current window locations to align displays, rather than using possibly-out-of-date cached location information. So it should "just work" now.

Comment 12 Julie 2015-01-29 02:48:21 UTC

(In reply to Jonathon Jongsma from comment #11)
> Sorry for the confusing wording. The user does not need to do anything. My
> comment simply means that virt-viewer itself will always use the current
> window locations to align displays, rather than using possibly-out-of-date
> cached location information. So it should "just work" now.

Thanks Jonathon.

Comment 14 errata-xmlrpc 2015-02-11 17:43:23 UTC

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.

https://rhn.redhat.com/errata/RHSA-2015-0197.html