similar to: [PATCH virt-v2v v2] v2v: -o libvirt: Remove cache=none.

Traditionally if you did live migration (KVM to KVM), you had to ensure that cache=none was set on all disks of the guest up front. This was because of quirks in how NFS works (I think the close-to-open consistency and the fact that during live migration both qemus have the file open), and we had to assume the worst case that a guest might be backed by NFS. Because of this when virt-v2v converts

There's a standardized libosinfo namespace for libvirt domain metadata. For now it supports the id of the OS only. However that is still a very helpful feature that is already supported in gnome-boxes and virt-manager (at least). The discussion happened here: https://www.redhat.com/archives/libosinfo/2018-September/msg00003.html So let's add the support to local and libvirt outputs.

Implement what old virt-v2v did (from lib/Sys/VirtConvert/Connection/LibVirtTarget.pm) Thanks: Tingting Zheng, Matthew Booth --- v2v/output_libvirt.ml | 118 ++++++++++++++++++++++++++++++++++++++++++++++--- v2v/output_libvirt.mli | 2 +- v2v/output_local.ml | 13 +++++- v2v/test-v2v-i-ova.xml | 5 ++- 4 files changed, 128 insertions(+), 10 deletions(-) diff --git

Does genid exist in oVirt ? If yes, is it already preserved or should it be a new patch ? On Thu, Jul 5, 2018 at 3:44 PM Richard W.M. Jones <rjones@redhat.com> wrote: > virt-v2v moves guests, it doesn't clone them. Therefore we should try > to preserve the VM Generation ID (genid) as much as possible. > > This has the ability to read the genid from VMware VMX files and

Where supported, pass the source CPU vendor, model and topology to the target hypervisor. For -i ova, we can get just cores per socket via a proprietary VMware extension to OVF. For -i libvirt and from virt-p2v, we can get all of these fields from the libvirt XML. For -o libvirt/local, we can preserve all of the information in the target XML. For -o glance, as far as I can tell from the

QEMU for x86 supports two machine types, "pc" (emulating the ancient Intel i440FX chipset originally used by the Pentium Pro), and "q35" (https://wiki.qemu.org/Features/Q35). Currently virt-v2v does not set any machine type, so libvirt or the target hypervisor will choose some default, probably i440fx. The latest advice from the QEMU and libvirt communities is not to rely on

Instead of storing the number of sockets, cores, and threads separately with the possibility to have each of them optional, store them together, so either we have a CPU topology or none. This slightly simplifies the handling of CPU topology. Most of the output modes/producers already considered a missing information of a topology as "1" when any of the other was specified, so the

On Tue, Jun 19, 2018 at 11:37:46AM +0100, Richard W.M. Jones wrote: > QEMU for x86 supports two machine types, "pc" (emulating the ancient > Intel i440FX chipset originally used by the Pentium Pro), and "q35" > (https://wiki.qemu.org/Features/Q35). > > Currently virt-v2v does not set any machine type, so libvirt or the > target hypervisor will choose some

virt-v2v moves guests, it doesn't clone them. Therefore we should try to preserve the VM Generation ID (genid) as much as possible. This has the ability to read the genid from VMware VMX files and libvirt XML (but note RHBZ#1598348). It can also write the genid to libvirt (‘-o libvirt’, ‘-o local’) and QEMU (‘-o qemu’). We are missing support currently for all OVF-based formats (hence ‘-i

QEMU for x86 supports two machine types, "pc" (emulating the ancient Intel i440FX chipset originally used by the Pentium Pro), and "q35" (https://wiki.qemu.org/Features/Q35). Currently virt-v2v does not set any machine type, so libvirt or the target hypervisor will choose some default, probably i440fx. Newer versions of libvirt and QEMU will probably switch over to defaulting

Create the correct tags for <video> in the libvirtxml, so all the attributes are in a <model> child of <video> instead of <video> itself. Adapt the XML of test-v2v-i-ova to the different expected XML. --- v2v/output_libvirt.ml | 9 +++++---- v2v/test-v2v-i-ova.xml | 4 +++- 2 files changed, 8 insertions(+), 5 deletions(-) diff --git a/v2v/output_libvirt.ml

Some ova files have their ovf and other files inside a folder rather than at the root of the tarball. Consider the paths relative to the ovf file to cover this case too. --- v2v/Makefile.am | 1 + v2v/input_ova.ml | 3 +- v2v/test-v2v-i-ova-subfolders.sh | 88 ++++++++++++++++++++++++++++++++++++++++ 3 files changed, 91 insertions(+), 1 deletion(-) create mode

The previous code treated floppy disks and CD-ROMs as the same kind of thing, resulting in malformed libvirt XML. You would see the following error when importing a guest into libvirt: error: Failed to define domain from /tmp/v2vlibvirt063486.xml error: internal error: Invalid floppy device name: hdb because we incorrectly generated this bogus libvirt XML fragment: <disk

Create two temporary directories for all the files created during the virt-v2v run: 1) tmpdir, created as $TMPDIR/virt-v2v.XXXXXX, for all the small files 2) cachedir, created as $LIBGUESTFS_CACHEDIR/virt-v2v.XXXXXX, for the big files (e.g. disks) This way there is no need to manually schedule all the temporary files and directories for removal when the application quits. --- v2v/input_ova.ml

On Fri, Nov 23, 2018 at 12:39:44PM +0100, Martin Kletzander wrote: > There's a standardized libosinfo namespace for libvirt domain metadata. For now > it supports the id of the OS only. However that is still a very helpful feature > that is already supported in gnome-boxes and virt-manager (at least). > > The discussion happened here: > >

On Thursday 19 May 2016 17:26:12 Cédric Bosdonnat wrote: > Some ova files have their ovf and other files inside a folder rather > than at the root of the tarball. Consider the paths relative to the > ovf file to cover this case too. > --- Ouch... this standard is getting more funky implementations, I see... what is the producer of this kind of ova? A couple of notes below. >

There's a standardized libosinfo namespace for libvirt domain metadata. For now it supports the id of the OS only. However that is still a very helpful feature that is already supported in gnome-boxes and virt-manager (at least). The discussion happened here: https://www.redhat.com/archives/libosinfo/2018-September/msg00003.html So let's add the support to local and libvirt outputs.

When writing the libosinfo metadata in the libvirt XML, use the newly added (in osinfo-db) ID for Windows Server 2019; sadly, this version of Windows has the same version as Windows Server 2016, so distinguish it by looking at its product name. --- v2v/create_libvirt_xml.ml | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/v2v/create_libvirt_xml.ml b/v2v/create_libvirt_xml.ml index

On Wed, Apr 05, 2017 at 02:43:11PM +0100, Richard W.M. Jones wrote: > When outputting libvirt XML, create virtio-rng, a memory balloon > device, and a pvpanic device, if the guest supports it. > --- > v2v/create_libvirt_xml.ml | 19 +++++++++++++++++++ > 1 file changed, 19 insertions(+) > > diff --git a/v2v/create_libvirt_xml.ml b/v2v/create_libvirt_xml.ml > index

CentOS 8 is represented with a simpler osinfo ID without a minor version. --- v2v/create_libvirt_xml.ml | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/v2v/create_libvirt_xml.ml b/v2v/create_libvirt_xml.ml index dbc24315..002ce2b8 100644 --- a/v2v/create_libvirt_xml.ml +++ b/v2v/create_libvirt_xml.ml @@ -43,9 +43,14 @@ let get_osinfo_id = function i_major_version