Red Hat Bugzilla – Bug 1283202
RFE: backport max monitor limitation from spice-server upstream
Last modified: 2017-08-01 12:06:08 EDT
Description of problem:
This is to support https://bugzilla.redhat.com/show_bug.cgi?id=1075798 in newer RHEL version.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Try to set monitor number from interface (like virt-manager).
2. Open spice client
3. Try to open more monitors than configured number
Displays are opened without problems.
Failure and client should not even prompt for such setting.
Needs to backport to spice-serve commit 6d4e58f70ddf9c178e2cc20dd5a3ade8b95fd142. Better to move to spice-server 0.12.6.
Still waiting Desktop team to sort out libvirt issue.
For me it is not fixed. If I set number of heads=1 in libvirt xml file, the number of displays can be increased up to four in remote-viewer menu (rhel 7.3 VM). Windows 7 VM works correctly.
host/client rhel 7.3 nightly:
client rhel 7.3 nightly
(In reply to Radek Duda from comment #4)
> For me it is not fixed. If I set number of heads=1 in libvirt xml file, the
> number of displays can be increased up to four in remote-viewer menu (rhel
> 7.3 VM). Windows 7 VM works correctly.
> host/client rhel 7.3 nightly:
What is the libvirt version? According to the bug 1283207 it should be fixed in libvirt-2.0.0-1.el7
> client rhel 7.3 nightly
Fairly sure QEMU is at fault https://bugzilla.redhat.com/show_bug.cgi?id=1283198#c7
ok Christophe I will test it later
I cannot verify this.
tested on client:
<model type='qxl' ram='262144' vram='262144' vgamem='65536' heads='1' primary='yes'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
in xml config file of VM guest
I can always open more displays than 1.
Should this really be FailedQE/ASSIGNED? I understand your testing fails, but I have no idea whether what I raised in comment #6 has been solved. Ie I don't think all the pieces are fully in place yet, but it's not necessarily broken on the spice-server side.
There are basically two options:
1) as not all pieces are obviously in we are not able fully test. So in theory is should be postponed until all bits are in.
2) we can verify as sanity only and fully test when qemu get fixed
1) is more clean approach from QE POW, on the other hand 2) is more likely from Devel POW. The only concern I have is that if we mark this bz sanity only there is very high possibility we won't test it later.
I would advice to ask in bz1283198 for status and emphasize it is a blocking issue for this bz.
> Additional info:
> Needs to backport to spice-serve commit
> 6d4e58f70ddf9c178e2cc20dd5a3ade8b95fd142. Better to move to spice-server
Backport isn't going to fly, breaks symbol versioning.
spice-server needs a rebase to 0.12.6 or newer.
moving to 7.4, we need 1283198 to be fixed as well
Should be part of the rebased spice-0.12.8-1.el7
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.