5:22 p.m.
Hi all, here's an overview of some virt happenings from the past month.
Fedora 19
=========
The feature freeze came and went, and some proposed virt features were dropped
from the list for not being fully testable in time:
https://fedoraproject.org/wiki/Features/Virt_Device_Failover
https://fedoraproject.org/wiki/Features/MQ_virtio_net
https://fedoraproject.org/wiki/Features/NetworkManagerBridging
https://fedoraproject.org/wiki/Features/oVirtEngine_3.2
So the final list is:
https://fedoraproject.org/wiki/Features/Virtio_RNG
https://fedoraproject.org/wiki/Features/Virt_Storage_Migration
https://fedoraproject.org/wiki/Features/QXLKMSSupport
https://fedoraproject.org/wiki/Features/OpenStack_Grizzly
However, the dropped features may still end up in the final F19 release, they
just aren't being marketed.
The next major bit is the Alpha release in a couple weeks, so keep an eye out
for the announcement, and give it a spin.
https://fedoraproject.org/wiki/Releases/19/Schedule
- April 16: Alpha Release
- May 21: Beta Release
- May 28: Virt Test Day
- June 25: Final Release
Libvirt 1.0.4 release
=====================
Libvirt 1.0.4 was released yesterday, April 1st. The feature list is:
qemu: support passthrough for iscsi disks (Paolo Bonzini),
various S390 improvements (Viktor Mihajlovski),
various LXC bugs fixes and improvements (Daniel P. Berrange, Gao feng),
Add API for thread cancellation (Daniel P. Berrange)
This is heading for F19 and rawhide. Additionally, a 0.10.2.4 maintenance
release is heading for F18.
Wiki page for using UEFI with KVM
=================================
I've added a wiki page describing how to use UEFI with KVM, specifically
targeted at testing Fedora's secureboot support:
https://fedoraproject.org/wiki/Testing_secureboot_with_KVM
Despite the secureboot focus, it's still a useful primer for trying out UEFI
and KVM in general. The page also documents my understanding of the licensing
issues that prevent us from distributing edk2/ovmf as a proper Fedora package.
Give it a spin and report here or bugzilla if you hit any issues.
Switch to seavgabios over vgabios
=================================
Qemu in F19/rawhide is now using seavgabios over the traditional vgabios
package. Motivation was that seabios has a much more active community, and the
new code is fair less terrifying :)
Similar to past switches like gpxe->ipxe this should be transparent to the
user, but if you hit any issues, please file a bug.
Bug stats
=========
Bug count on Mar 04 2013 : 247
Bug count on Apr 02 2013 : 244
There was actually a lot of bug activity this month, though the numbers don't
show it. If I generated this number yesterday morning it would have been much
higher :)
There was quite a number of bugs filed about handling 64bit arm support for
older configure scripts, which is responsible for the bug increases in the
more obscure packages.
By release:
* Fedora 17 : 87
* Fedora 18 : 115
* Fedora 19 : 8
* Fedora rawhide : 34
By package:
* bochs : 1
* gnome-boxes : 21
* gtk-vnc : 1
* ipxe : 3
* libcacard : 1
* libguestfs : 5
* libosinfo : 7
* libseccomp : 1
* libvirt : 43
* libvirt-cim : 2
* libvirt-glib : 2
* libvirt-qmf : 1
* libvirt-sandbox : 19
* netcf : 5
* python-virtinst : 2
* qemu : 56
* spice-gtk : 4
* spice-vdagent : 2
* usbredir : 1
* virt-dmesg : 2
* virt-manager : 21
* virt-top : 1
* virt-v2v : 3
* virt-viewer : 6
* virt-what : 2
* virtio-win : 4
* xen : 6
* xorg-x11-drv-cirrus: 4
* xorg-x11-drv-qxl : 18
By status:
* ASSIGNED : 21
* MODIFIED : 11
* NEW : 203
* ON_QA : 6
* POST : 3
Bugs of note
============
https://bugzilla.redhat.com/show_bug.cgi?id=872162
An LVM volume created through virt-manager with default settings doesn't work.
This has been an ongoing saga since around F18 alpha timeframe. virt-manager
wanted to create a 'sparse' lvm volume. Originally libvirt ignored this and
gave us a working result. When libvirt began actually handling it, turns out
that lvm sparse volumes don't auto allocate like a sparse file, which upset
common VM usage. virt-manager fixed its default, so this shouldn't happen again.
Thanks,
Cole