Created attachment 1766345 [details]
virt-v2v-convert-win2019-to-local.log

Starting some time in RHEL 9 we will just use standard VGA.  See:
https://bugzilla.redhat.com/show_bug.cgi?id=1961107#c1

Bulk update: Move RHEL-AV bugs to RHEL8.

Per comment 2, clone unnecessary.

Leaving this open for now since I think it's a real bug.  Even though
we will deprecate QXL in RHEL 9, there should be a fix in 8.6 if we
can find the names of the QXL drivers on the virtio-win disk and
get virt-v2v to match those names correctly.

The attachment from comment#1 contains:

> libguestfs: trace: v2v: exists "/Windows/Drivers/VirtIO/qxl.inf"
> 
> guestfsd: <= exists (0x24) request length 76 bytes
> guestfsd: => exists (0x24) took 0.00 secs
> libguestfs: trace: v2v: exists = 0
> virt-v2v: warning: there is no QXL driver for this version of Windows (10.0 
> x86_64).  virt-v2v looks for this driver in 
> /usr/share/virtio-win/virtio-win.iso
> 
> The guest will be configured to use a basic VGA display driver.

According to the log, there is no check for "qxldod.inf", which is what the virtio-win RPM / ISO provide.

However, this issue has been fixed upstream meanwhile. Originally in commit 11d1f3cd6878 ("v2v: windows: Allow qxldod.inf as synonym for qxl.inf", 2021-04-27), on the master branch.

Then it has been cherry-picked to the rhel-8.6.0 branch even, namely as commit 575cb719ceb56 (shared with the rhel-av-8.5.0 branch).

The commit message references bug 1926102, which was filed for RHEL-8.4, and is currently in status CLOSED ERRATA. The Fixed-In-Version field on bug 1926102 says "virt-v2v-1.42.0-11.el8".

The present bug was also filed for RHEL-8.4, specifically against "virt-v2v-1.42.0-9.module+el8.4.0+9561+069bb9c1.x86_64". Note the Release part of the NVR:

  virt-v2v-1.42.0-11.el8                                 = the fix for bug 1926102
  virt-v2v-1.42.0-9.module+el8.4.0+9561+069bb9c1.x86_64  = the basis for reporting bug 1943203 (= this bug)

This means that the present bug is a duplicate of 1926102; it was reported after bug 1926102 was reported, but before bug 1926102 was fixed.

*** This bug has been marked as a duplicate of bug 1926102 ***

This bug is for RHEL-8, and bug 1926102 is for RHEL-AV-8, so I think closing this one as a duplicate of 1926102 is not OK.

Here's why:

(1) The latest tag in the "module-virt-rhel-8" series (not "module-virt-av-8" series) is:

$ brew list-tags | grep module-virt-rhel-8
...
module-virt-rhel-8060020220118011525-d63f516d
...

(2) The latest build of virt-v2v in this tag is:

brew latest-build \
  module-virt-rhel-8060020220118011525-d63f516d \
  virt-v2v

virt-v2v-1.42.0-18.module+el8.6.0+13447+4b5d0856 ...

Available at <https://brewweb.engineering.redhat.com/brew/buildinfo?buildID=1805851>.

The changelog contains (note the transition from virt-v2v-1.42.0-9 to virt-v2v-1.42.0-11):

* Tue Apr 27 2021 Richard W.M. Jones <rjones> - 1:1.42.0-11
- v2v: windows: Allow qxldod.inf as synonym for qxl.inf <------------------------ see this here
  resolves: rhbz#1926102
- v2v: Increase required free space in Windows to 100 MB
  resolves: rhbz#1949147
- docs: Document how to remove "Out of HTTP sessions" limit
- v2v: Disable readahead for VMware curl sources too
  resolves: rhbz#1848862
- v2v: Allow output to block devices
  resolves: rhbz#1868690
- docs: -o openstack: Clarify name of file containing OpenStack auth
  resolves: rhbz#1871754
- docs: UEFI guest conversion to -o openstack is supported
  resolves: rhbz#1872100
- v2v: Turn pnp_wait.exe warning into a debug message
  resolves: rhbz#1903960
- v2v: windows: Fix schtasks /SD parameter
  resolves: rhbz#1895323

Therefore, this BZ should indeed be in MODIFIED state (to be checked separately by virt-tools QE). Testing should target the above-noted latest build for RHEL-8 (not AV).

Yup, the "stream-rhel-rhel-8.6.0" branch in dist-git currently stands at commit 8e01a6661313, corresponding to epoch-version-release 1:1.42.0-18%{?dist}. The fix for this particular bug (= upstream 575cb719ceb56) was included in dist-git commit f492feabe0ad, corresponding to build 1:1.42.0-11%{?dist}.

I'm not actually sure if it's right to mark the bug as duplicate or not
(perhaps one for Danilo to answer).  But I will note that in RHEL 8.6,
the virt module RHEL AV and RHEL 8 streams converge.  We only need to fix
things once.

The bug has been fixed and passed auto jobs,video type has been set as qxl after v2v converting win10, win2016 and win2019 guests:https://libvirt-jenkins.rhev-ci-vms.eng.rdu2.redhat.com/view/v2v/view/RHEL-8.6/job/v2v-RHEL-8.6-runtest-x86_64-acceptance-libvirt/lastSuccessfulBuild/testReport/rhel/convert_vm_to_libvirt/

So set Veried:Tested filed.

Tested the bug with below builds:
virtio-win-1.9.22-1.el8.noarch
qemu-kvm-6.2.0-4.module+el8.6.0+13919+adb438db.x86_64
libvirt-8.0.0-1.module+el8.6.0+13888+55157bfb.x86_64
virt-v2v-1.42.0-18.module+el8.6.0+13447+4b5d0856.x86_64
libguestfs-winsupport-8.6-1.module+el8.6.0+12861+13975d62.x86_64

Convert win10, win2016 and win2019 guests from esxi 7.0 to local kvm, all guests have video type set as qxl in xml file after conversion, then start the guests and found qxl drivers have been installed successfully in guests.

Based on the above comments, move the bug to VERIFIED.

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 (Moderate: virt:rhel and virt-devel:rhel security, bug fix, and enhancement update), 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/RHSA-2022:1759