Bug 1461684

Summary: virt-manager should rebuild dom capabilities when chipset updated
Product: Red Hat Enterprise Linux 7 Reporter: Xiaodai Wang <xiaodwan>
Component: virt-managerAssignee: Pavel Hrdina <phrdina>
Status: CLOSED ERRATA QA Contact: Virtualization Bugs <virt-bugs>
Severity: low Docs Contact:
Priority: low    
Version: 7.4CC: juzhou, kuwei, mxie, mzhan, phrdina, tzheng
Target Milestone: rc   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
Whiteboard:
Fixed In Version: virt-manager-1.4.3-1.el7 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-04-10 11:40:46 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:
Attachments:
Description Flags
Screenshot-1 none

Description Xiaodai Wang 2017-06-15 07:39:49 UTC
Description of problem:
virt-manager should rebuild dom capabilities when chip-set updated

Version-Release number of selected component (if applicable):
virt-manager-1.4.1-7.el7.noarch

How reproducible:
100%

Steps to Reproduce:
1. Create a new vm, and check "Customize configuration before install" at step 5 of 5, then click finish.
2. In the customize window, modify the chipset to "q35" and firmware to UEFI, then click Apply button.
3. Select the disk and check the disk bus of the disk.

Actual results:
"IDE" is still be listed in Disk bus.

Expected results:
"IDE" should be removed when chip-set is Q35.

Additional info:

Comment 1 Pavel Hrdina 2017-09-19 08:42:23 UTC
Upstream patch posted:

https://www.redhat.com/archives/virt-tools-list/2017-September/msg00101.html

Comment 2 Pavel Hrdina 2017-09-20 08:29:39 UTC
Upstream commit:

commit fd420fdacd2caa341e6c47cf56c9d8d1b9feee53
Author: Pavel Hrdina <phrdina>
Date:   Tue Sep 19 10:40:09 2017 +0200

    domain: invalidate domain caps if machine type is changed

Comment 4 zhoujunqin 2017-09-22 08:10:55 UTC
I try to verify this bug with new build:
virt-manager-1.4.3-1.el7.noarch

Steps:
1. Create a new vm with 'Network Install' method, and check "Customize configuration before install" at step 5 of 5, then click finish.
2. In the customize window, modify the chipset to "q35" and firmware to UEFI, then click Apply button.
3. Select the disk and check the disk bus of the disk.

Result:
1) After step3, disk bus type shows: virtio/usb/sata/scsi, no 'ide' lists.
2) If i click 'Add Hardware' button, then i checked bus type for CDROM device is:  sata/scsi.

But Pavel, i also have a question here want to confirm with you, thanks.
When I create a vm with Local ISO image method, then repeat step 1&2,
I find disk bus type keeps 'IDE' for cdrom after step2, details please see screenshot-1, but in disk bus list there is no 'IDE' listing here.

If i click 'Begin installation', it will failed with error:
Unable to complete install: 'unsupported configuration: IDE controllers are unsupported for this QEMU binary or machine type'
I need to change cdrom from ide to sata or scsi.

I'm not sure whether we can accept it or not, thanks.

Comment 5 zhoujunqin 2017-09-22 08:15:09 UTC
Created attachment 1329451 [details]
Screenshot-1

Comment 6 Pavel Hrdina 2017-09-22 08:42:05 UTC
I would create a new BUG for that.  This will require some more coding because it may affect more devices, not only disks.

Comment 7 zhoujunqin 2017-09-25 02:42:33 UTC
(In reply to Pavel Hrdina from comment #6)
> I would create a new BUG for that.  This will require some more coding
> because it may affect more devices, not only disks.

Ok, Pavel, after you file that bug, please leave bug id here, thanks.

Comment 8 Pavel Hrdina 2017-09-26 08:37:15 UTC
Now when I'm thinking about it a little bit more I'm not sure, whether we should change the disk bus and other default values when the chipset is changed.  User can decide to switch it back or would like to use that specific configuration.  With the current behavior the error message will tell them that this configuration isn't possible and they can decide to change the chipset back or update the affected devices.

I would rather keep the current behavior with the error message.

Comment 9 zhoujunqin 2017-09-26 10:00:31 UTC
(In reply to Pavel Hrdina from comment #8)
> Now when I'm thinking about it a little bit more I'm not sure, whether we
> should change the disk bus and other default values when the chipset is
> changed.  User can decide to switch it back or would like to use that
> specific configuration.  With the current behavior the error message will
> tell them that this configuration isn't possible and they can decide to
> change the chipset back or update the affected devices.
> 
> I would rather keep the current behavior with the error message.

Hi Pavel
Don't worry, it works for me.
For the error message is clear enough for user.

So I move this bug from ON_QA to VERIFIED status, thanks.

Comment 12 errata-xmlrpc 2018-04-10 11:40:46 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://access.redhat.com/errata/RHEA-2018:0726