]> git.proxmox.com Git - qemu-server.git/log
qemu-server.git
9 years agodrive-mirror : wait that busy eq false before block-job-complete
Alexandre Derumier [Fri, 7 Nov 2014 09:10:48 +0000 (10:10 +0100)]
drive-mirror : wait that busy eq false before block-job-complete

When the drive-mirror is at 100%, and write occurs, the busy flag can change from false->true

- 100% no new writes

transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false

- 100% new writes

transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: true
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: true
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: true

- 100% no new writes

transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false
transferred: 1073741824 bytes remaining: 0 bytes total: 1073741824 bytes progression: 100.00 % busy: false

So, we need to check that busy is false before doing the block-job-complete.

Also, we force the vm to pause, if it's busy for more than 120s when drive-mirror is at 100%

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoblock-job-cancel : wait that jobs is really finished
Alexandre Derumier [Fri, 7 Nov 2014 09:10:47 +0000 (10:10 +0100)]
block-job-cancel : wait that jobs is really finished

block-job-cancel is async, we need to check that job is really finished

before try to free the volume

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.3-1
Dietmar Maurer [Mon, 13 Oct 2014 08:36:23 +0000 (10:36 +0200)]
bump version to 3.3-1

9 years agoenable write zeroes optimisations
Alexandre Derumier [Mon, 13 Oct 2014 07:45:30 +0000 (09:45 +0200)]
enable write zeroes optimisations

This enable write zeroes optimisation inside guest.

(I have tested them with qcow2,raw (ext4|xfs) , zfs|iscsi and nfs too.

No impact if the block driver don't support it yet (like rbd)

https://lists.nongnu.org/archive/html/qemu-devel/2014-04/msg00009.html

> a) mkfs.ext4 -E lazy_itable_init=0,lazy_journal_init=0 /dev/vdX
>
> QCOW2         [off]     [on]     [unmap]
> -----
> runtime:       14secs    1.1secs  1.1secs
> filesize:      937M      18M      18M
>
> iSCSI         [off]     [on]     [unmap]
> ----
> runtime:       9.3s      0.9s     0.9s
>
> b) dd if=/dev/zero of=/dev/vdX bs=1M oflag=direct
>
> QCOW2         [off]     [on]     [unmap]
> -----
> runtime:       246secs   18secs   18secs
> filesize:      51G       192K     192K
> throughput:    203M/s    2.3G/s   2.3G/s
>
> iSCSI*        [off]     [on]     [unmap]
> ----
> runtime:       8mins     45secs   33secs
> throughput:    106M/s    1.2G/s   1.6G/s
> allocated:     100%      100%     0%

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoadd iothread/dataplane support
Alexandre Derumier [Fri, 10 Oct 2014 02:05:44 +0000 (04:05 +0200)]
add iothread/dataplane support

new config option:

iothread: 1|0

This enable iothread/dataplane support, to improve io performance on fast storages

Currently block jobs don't work yet, it's planned for qemu 2.2.
So it's better to not expose yet this option in gui.

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-35
Dietmar Maurer [Wed, 17 Sep 2014 13:53:02 +0000 (15:53 +0200)]
bump version to 3.1-35

9 years agofix bug #542: return VMID as integer
Dietmar Maurer [Wed, 17 Sep 2014 13:51:47 +0000 (15:51 +0200)]
fix bug #542: return VMID as integer

9 years agobump version to 3.1-34
Dietmar Maurer [Mon, 1 Sep 2014 09:36:07 +0000 (11:36 +0200)]
bump version to 3.1-34

9 years agovm_devices_list : also list block devices
Alexandre Derumier [Fri, 29 Aug 2014 13:04:15 +0000 (15:04 +0200)]
vm_devices_list : also list block devices

This allow scsi disk to be plug|unplug

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoCurrently,if we don't have a "machine" option in running config, and we take a vmstat...
Dietmar Maurer [Fri, 29 Aug 2014 09:27:27 +0000 (11:27 +0200)]
Currently,if we don't have a "machine" option in running config, and we take a vmstate snapshot

the machine option is write in the snapshot (ok), but also in the running config (bad)

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
9 years agoallow hotplug of virtio-scsi disks
Alexandre Derumier [Wed, 27 Aug 2014 13:48:30 +0000 (15:48 +0200)]
allow hotplug of virtio-scsi disks

It was an old protection, it's working fine now

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-33
Dietmar Maurer [Tue, 26 Aug 2014 07:24:29 +0000 (09:24 +0200)]
bump version to 3.1-33

9 years agoclone_vm: auto generate new uuid
Dietmar Maurer [Tue, 26 Aug 2014 07:20:09 +0000 (09:20 +0200)]
clone_vm: auto generate new uuid

9 years agobump version to 3.1-32
Dietmar Maurer [Wed, 20 Aug 2014 10:21:28 +0000 (12:21 +0200)]
bump version to 3.1-32

9 years agoadd Broadwell cpu model
Alexandre Derumier [Mon, 18 Aug 2014 13:56:35 +0000 (15:56 +0200)]
add Broadwell cpu model

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-31
Dietmar Maurer [Wed, 13 Aug 2014 04:17:31 +0000 (06:17 +0200)]
bump version to 3.1-31

9 years agogenerate spice devices cmd line after pci bridge
Alexandre Derumier [Tue, 12 Aug 2014 23:04:31 +0000 (01:04 +0200)]
generate spice devices cmd line after pci bridge

we should push to $devices array instead $cmd array,

because pci bridges need to be create before spice devices

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-30
Dietmar Maurer [Wed, 6 Aug 2014 07:41:43 +0000 (09:41 +0200)]
bump version to 3.1-30

9 years agobump max hostpci to 4
Alexandre Derumier [Wed, 6 Aug 2014 05:39:33 +0000 (07:39 +0200)]
bump max hostpci to 4

user need to passthough 3 devices here
http://forum.proxmox.com/threads/19134-KVM-passtrought-PciExpress-3-card

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agovga=none if x-vga passthrough is enabled
Alexandre Derumier [Wed, 6 Aug 2014 05:27:29 +0000 (07:27 +0200)]
vga=none if x-vga passthrough is enabled

we need to disable virtual vga card if we passthrough a physical gpu

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-29
Dietmar Maurer [Tue, 29 Jul 2014 04:53:27 +0000 (06:53 +0200)]
bump version to 3.1-29

9 years agovm_stop: do not use ha commands if $migratedfrom is set
Dietmar Maurer [Tue, 29 Jul 2014 04:51:02 +0000 (06:51 +0200)]
vm_stop: do not use ha commands if $migratedfrom is set

9 years agobump version to 3.1-28
Dietmar Maurer [Thu, 24 Jul 2014 04:52:43 +0000 (06:52 +0200)]
bump version to 3.1-28

9 years agodisable kvm cpu signature if x-vga is enabled
Alexandre Derumier [Wed, 23 Jul 2014 12:38:00 +0000 (14:38 +0200)]
disable kvm cpu signature if x-vga is enabled

see
http://git.qemu.org/?p=qemu.git;a=commit;h=f522d2acc549dd11f495048330aa5f3f424a7dfa

last nvdia drivers don't install in kvm machine if they detect kvm signature.

This patch hide kvm signature in cpuflags (but don't disable kvm)

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-27
Dietmar Maurer [Wed, 23 Jul 2014 04:12:40 +0000 (06:12 +0200)]
bump version to 3.1-27

9 years agoReset device only if has_fl_reset is defined
Kamil Trzciński [Sun, 29 Jun 2014 19:41:37 +0000 (21:41 +0200)]
Reset device only if has_fl_reset is defined

Signed-off-by: Kamil Trzciński <ayufan@ayufan.eu>
9 years agobump version to 3.1-26
Dietmar Maurer [Thu, 17 Jul 2014 07:29:01 +0000 (09:29 +0200)]
bump version to 3.1-26

9 years agosnapshot_delete: reorder parent "connection" after prepare
Stefan Priebe [Fri, 4 Jul 2014 08:25:49 +0000 (10:25 +0200)]
snapshot_delete: reorder parent "connection" after prepare

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
9 years agoAPI2: remove require full parameter for snapshot cloning
Stefan Priebe [Fri, 4 Jul 2014 08:25:48 +0000 (10:25 +0200)]
API2: remove require full parameter for snapshot cloning

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
9 years agopass snapname to PVE::Storage::vdisk_clone
Stefan Priebe [Fri, 4 Jul 2014 08:25:47 +0000 (10:25 +0200)]
pass snapname to PVE::Storage::vdisk_clone

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
9 years agobump version to 3.1-25
Dietmar Maurer [Wed, 16 Jul 2014 10:48:10 +0000 (12:48 +0200)]
bump version to 3.1-25

9 years agoallow resize of virtio windows boot disk
Alexandre Derumier [Wed, 16 Jul 2014 08:36:19 +0000 (10:36 +0200)]
allow resize of virtio windows boot disk

virtio-win-0.1-74  have fixed the resize bug of virtio boot disk

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoauto generate uuid on VM create
Dietmar Maurer [Thu, 26 Jun 2014 09:51:52 +0000 (11:51 +0200)]
auto generate uuid on VM create

Other VM environments also generates an UUID by default.

9 years agorequire 'VM.Config.HWType' priviledge to change new smbios1 option
Dietmar Maurer [Thu, 26 Jun 2014 09:17:10 +0000 (11:17 +0200)]
require 'VM.Config.HWType' priviledge to change new smbios1 option

9 years agobump version to 3.1-24
Dietmar Maurer [Thu, 26 Jun 2014 09:13:50 +0000 (11:13 +0200)]
bump version to 3.1-24

9 years agonew option smbios1: specify SMBIOS type 1 fields (uuid, ...)
Dietmar Maurer [Thu, 26 Jun 2014 09:12:25 +0000 (11:12 +0200)]
new option smbios1: specify SMBIOS type 1 fields (uuid, ...)

9 years agobump version to 3.1-23
Dietmar Maurer [Wed, 25 Jun 2014 07:56:24 +0000 (09:56 +0200)]
bump version to 3.1-23

9 years agovncproxy: remove check if VM is running
Dietmar Maurer [Wed, 25 Jun 2014 07:54:23 +0000 (09:54 +0200)]
vncproxy: remove check if VM is running

Because this does not work when VM is on other cluster node.

9 years agobump version to 3.1-22
Dietmar Maurer [Wed, 25 Jun 2014 07:31:59 +0000 (09:31 +0200)]
bump version to 3.1-22

9 years agoadd pci multifunction unbind support
Alexandre Derumier [Mon, 23 Jun 2014 15:41:55 +0000 (17:41 +0200)]
add pci multifunction unbind support

we need to unbind each functions of a multifunction pci device

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoadd pci multifunction support
Alexandre Derumier [Mon, 23 Jun 2014 15:41:54 +0000 (17:41 +0200)]
add pci multifunction support

multifunction device should be define without the .function

hostpci0: 00:00

example
-------
if 00:00.0
   00:00.1
   00:00.2

exists,

then we generate the multifunction devices

-device (pci-assign|vfio-pci),host=00:00.0,id=hostpci0.0,bus=...,addr=0x0.0,multifunction=on
-device (pci-assign|vfio-pci),host=00:00.1,id=hostpci0.1,bus=...,addr=0x0.1
-device (pci-assign|vfio-pci),host=00:00.2,id=hostpci0.2,bus=...,addr=0x0.2

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoadd pcie and x-vga passthrough
Alexandre Derumier [Mon, 23 Jun 2014 15:41:53 +0000 (17:41 +0200)]
add pcie and x-vga passthrough

hostpci0:  .....,x-vga=on,pcie=1

x-vga require kernel 3.10 with vfio-vga support enable
if x-vga=on, we force vfio-pci device

pcie=1 choose the pciexpress bus (need q35 machine model)

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agovncproxy: add check if VM is running
Dietmar Maurer [Wed, 25 Jun 2014 05:20:23 +0000 (07:20 +0200)]
vncproxy: add check if VM is running

To provide better eror message for console task.

9 years agobump version to 3.1-21
Dietmar Maurer [Tue, 24 Jun 2014 15:43:54 +0000 (17:43 +0200)]
bump version to 3.1-21

9 years agoprotect websocket API with vncticket
Dietmar Maurer [Tue, 24 Jun 2014 14:10:30 +0000 (16:10 +0200)]
protect websocket API with vncticket

9 years agofix a typo
Dietmar Maurer [Tue, 24 Jun 2014 05:44:17 +0000 (07:44 +0200)]
fix a typo

9 years agobump version to 3.1-20
Dietmar Maurer [Wed, 18 Jun 2014 10:46:02 +0000 (12:46 +0200)]
bump version to 3.1-20

9 years agovncwebsocket: do not proxy connection
Dietmar Maurer [Wed, 18 Jun 2014 10:44:46 +0000 (12:44 +0200)]
vncwebsocket: do not proxy connection

9 years agobump version to 3.1-19
Dietmar Maurer [Wed, 18 Jun 2014 09:05:05 +0000 (11:05 +0200)]
bump version to 3.1-19

9 years agodo not use novnc wsproxy
Dietmar Maurer [Wed, 18 Jun 2014 09:02:57 +0000 (11:02 +0200)]
do not use novnc wsproxy

Instead, we use new HTTPServer features.

9 years agoadded support for vfio-pci passthrough
Stefan Priebe [Thu, 13 Feb 2014 20:12:29 +0000 (21:12 +0100)]
added support for vfio-pci passthrough

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
9 years agocleanup previous patch
Dietmar Maurer [Wed, 18 Jun 2014 04:54:45 +0000 (06:54 +0200)]
cleanup previous patch

9 years agoenable q35 machine support
Alexandre Derumier [Mon, 26 May 2014 07:49:56 +0000 (09:49 +0200)]
enable q35 machine support

q35 use pcie.0 root by default. so currently we can't start machine model q35.

we need to add 3 pci-bridge pci.0, pci.1, pci.2, to handle our devices.

pcie.0 does not support hotplug. so pci-bridge are defined at startup.

I use an pve-q35.cfg (mostly the same than q35-chipset.cfg from qemu docs).

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-18
Dietmar Maurer [Tue, 17 Jun 2014 07:00:24 +0000 (09:00 +0200)]
bump version to 3.1-18

9 years agomigration : add setup state
Alexandre Derumier [Tue, 17 Jun 2014 05:44:05 +0000 (07:44 +0200)]
migration : add setup state

since qemu 1.5, they are a new migration state : "setup"

it's mainly use for rdma migration, but slow vm can it see and hang on migration

http://git.qemu.org/?p=qemu.git;a=commit;h=3b6959506831193f37cc830c8e111b437c0d1380

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agodepend on novnc-pve
Dietmar Maurer [Tue, 17 Jun 2014 06:54:23 +0000 (08:54 +0200)]
depend on novnc-pve

9 years agocleanup VNC websocket patch
Dietmar Maurer [Tue, 17 Jun 2014 06:02:43 +0000 (08:02 +0200)]
cleanup VNC websocket patch

Pass ticket via ENV{LC_PVE_TICKET} to vncproxy. Also remove 'unsecure' option, because
this is implied if we want a websocket connection.

9 years agoAPI2/Qemu: add unsecure and websocket options to vncpoxy also set qemu vnc server...
Stefan Priebe [Tue, 3 Jun 2014 08:22:00 +0000 (10:22 +0200)]
API2/Qemu: add unsecure and websocket options to vncpoxy also set qemu vnc server properties on the fly

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
9 years agoadd virtio-net multiqueue support
Alexandre Derumier [Tue, 10 Jun 2014 06:30:31 +0000 (08:30 +0200)]
add virtio-net multiqueue support

this a new option queue=(\d+) to net interface

Allow to use more than 1 cpu for network stream, so this can improve network bandwidth,
when vhost-net cpu is the bottleneck

http://www.linux-kvm.org/page/Multiqueue#Enable_MQ_feature

-netdev tap,vhost=on,queues=N -device virtio-net-pci,mq=on,vectors=2N+2

host requirement
----------------
this require host kernel >= 3.8 (or qemu die at start)

linux guest requirement
-----------------------
kernel >= 3.8

manual enabling multiqueue

windows guest requierement
--------------------------
recent virtio-net driver

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoadd option iscsi only once
Dietmar Maurer [Sat, 17 May 2014 07:14:58 +0000 (09:14 +0200)]
add option iscsi only once

We simply add option iscsi if we have an initiator name. So we
never add this option multiple times, and it works with hotplug
in case someone plugs an 'iscsi:' drive later.

9 years agocleanup previous commit - use IO::File
Dietmar Maurer [Sat, 17 May 2014 07:07:18 +0000 (09:07 +0200)]
cleanup previous commit - use IO::File

9 years agoadd initiator-name to iscsi drives if configured
Michael Rasmussen [Tue, 13 May 2014 01:10:40 +0000 (03:10 +0200)]
add initiator-name to iscsi drives if configured

Signed-off-by: Michael Rasmussen <mir@datanom.net>
9 years agoprint_net : add firewall option
Alexandre Derumier [Fri, 9 May 2014 09:18:58 +0000 (11:18 +0200)]
print_net : add firewall option

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agodo not ignore errors on tap_unplug
Dietmar Maurer [Thu, 8 May 2014 13:45:25 +0000 (15:45 +0200)]
do not ignore errors on tap_unplug

9 years agoadd pve-bridgedown script
Alexandre Derumier [Wed, 7 May 2014 08:42:43 +0000 (10:42 +0200)]
add pve-bridgedown script

This allow to delete fwbr bridge, link interfaces, cleanup ovs
when vm is shutdown

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agoadd firewall option to qemu network interface
Alexandre Derumier [Wed, 7 May 2014 08:42:42 +0000 (10:42 +0200)]
add firewall option to qemu network interface

this allow to disable firewall for a specific interface

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
9 years agobump version to 3.1-17
Dietmar Maurer [Tue, 6 May 2014 09:28:42 +0000 (11:28 +0200)]
bump version to 3.1-17

9 years agodepend on pve-firewall, add firewall API for VMs
Dietmar Maurer [Tue, 6 May 2014 09:27:10 +0000 (11:27 +0200)]
depend on pve-firewall, add firewall API for VMs

10 years agoqmrestore: removed short timeout
Dietmar Maurer [Thu, 17 Apr 2014 08:37:46 +0000 (10:37 +0200)]
qmrestore: removed short timeout

10 years agoset RELEASE to 3.2
Dietmar Maurer [Thu, 17 Apr 2014 07:44:27 +0000 (09:44 +0200)]
set RELEASE to 3.2

10 years agoadd linefeed to warn message and add changelog for previous commit
Dietmar Maurer [Thu, 17 Apr 2014 07:29:46 +0000 (09:29 +0200)]
add linefeed to warn message and add changelog for previous commit

10 years agomove_disk : don't delete disk if used in a previous snasphot v2
Alexandre Derumier [Mon, 14 Apr 2014 11:52:23 +0000 (13:52 +0200)]
move_disk : don't delete disk if used in a previous snasphot v2

and set it as unused

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agobump version to 3.1-16
Dietmar Maurer [Mon, 14 Apr 2014 09:28:31 +0000 (11:28 +0200)]
bump version to 3.1-16

10 years agofix bug #502: allow creation of empty vma archives
Dietmar Maurer [Mon, 14 Apr 2014 09:21:01 +0000 (11:21 +0200)]
fix bug #502: allow creation of empty vma archives

10 years agodisable cpu enforce flag
Dietmar Maurer [Thu, 3 Apr 2014 07:39:54 +0000 (09:39 +0200)]
disable cpu enforce flag

Enforce needs kernel 3.10, so we do not use it for now.

10 years agoenforce cpu check
Alexandre Derumier [Thu, 27 Mar 2014 12:47:30 +0000 (13:47 +0100)]
enforce cpu check

enable check if host support all cpu flags configured for the guests

this avoid some bad setup like Opteron vcpu on a intel host for example,
and avoid some bad live migrations

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agoadd depend on libpve-access-control
Dietmar Maurer [Mon, 17 Mar 2014 09:53:15 +0000 (10:53 +0100)]
add depend on libpve-access-control

10 years agofix changelog
Dietmar Maurer [Mon, 10 Feb 2014 11:30:08 +0000 (12:30 +0100)]
fix changelog

10 years agoQemuMigrate: print migration xbzrle if enabled (has xbzrlecachesize) for whatever...
Stefan Priebe [Sun, 9 Feb 2014 09:05:48 +0000 (10:05 +0100)]
QemuMigrate: print migration xbzrle if enabled (has xbzrlecachesize) for whatever reason (bug qemu, bug pve, ...)

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
10 years agobump version to 3.1-15
Dietmar Maurer [Mon, 10 Feb 2014 07:07:06 +0000 (08:07 +0100)]
bump version to 3.1-15

10 years agoremove unneeded parentheses
Dietmar Maurer [Mon, 10 Feb 2014 07:03:50 +0000 (08:03 +0100)]
remove unneeded parentheses

10 years agosince qemu 1.7 xbzrle is AUTOMATICALLY enabled in the end of transfer - we need to...
Stefan Priebe [Sat, 8 Feb 2014 19:20:55 +0000 (20:20 +0100)]
since qemu 1.7 xbzrle is AUTOMATICALLY enabled in the end of transfer - we need to explicit disable all feature we do not want

Signed-off-by: Stefan Priebe <s.priebe@profihost.ag>
10 years agoadd changelog for last commit
Dietmar Maurer [Fri, 31 Jan 2014 05:52:46 +0000 (06:52 +0100)]
add changelog for last commit

10 years agoAdds the VM name to the title for the spice clent, making it easier to identify which...
Lindsay Mathieson [Thu, 16 Jan 2014 13:21:05 +0000 (23:21 +1000)]
Adds the VM name to the title for the spice clent, making it easier to identify which in window lists etc.

Signed-off-by: Lindsay Mathieson <lindsay.mathieson@gmail.com>
10 years agobump version to 3.1-14
Dietmar Maurer [Wed, 29 Jan 2014 05:49:45 +0000 (06:49 +0100)]
bump version to 3.1-14

10 years agodeactivate volume after clone
Dietmar Maurer [Wed, 29 Jan 2014 05:44:06 +0000 (06:44 +0100)]
deactivate volume after clone

To avoid lvm LVs to be active on several nodes.

10 years agowhite space cleanups
Dietmar Maurer [Tue, 14 Jan 2014 09:33:36 +0000 (10:33 +0100)]
white space cleanups

10 years agoadd cpu_hotplug
Alexandre Derumier [Tue, 7 Jan 2014 12:32:51 +0000 (13:32 +0100)]
add cpu_hotplug

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agoadd maxcpus config
Alexandre Derumier [Tue, 7 Jan 2014 12:32:50 +0000 (13:32 +0100)]
add maxcpus config

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agocleanup previous commit
Dietmar Maurer [Fri, 10 Jan 2014 12:09:19 +0000 (13:09 +0100)]
cleanup previous commit

10 years agomigration : enable auto-converge capability v2
Alexandre Derumier [Thu, 9 Jan 2014 09:58:48 +0000 (10:58 +0100)]
migration : enable auto-converge capability v2

This reduce guest cpu speed if dirtied bytes is 50% more than the approx.amount of bytes that just got transferred since the last time we were in this routine.

qemu commit :
http://git.qemu.org/?p=qemu.git;a=commit;h=bde1e2ec2176c363c1783bf8887b6b1beb08dfee

tested with "stress -m 2 -c 2" under debian

without autoconvergence : downtime 12s - duration 12min
with autoconvergence : downtime 2s - duration 4min

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agobump version to 3.1-13
Dietmar Maurer [Fri, 13 Dec 2013 10:46:25 +0000 (11:46 +0100)]
bump version to 3.1-13

10 years agoadd more options for pci passthrough
Dietmar Maurer [Fri, 13 Dec 2013 10:43:05 +0000 (11:43 +0100)]
add more options for pci passthrough

10 years agobump version to 3.1-12
Dietmar Maurer [Tue, 10 Dec 2013 09:49:55 +0000 (10:49 +0100)]
bump version to 3.1-12

10 years agocode cleanup
Dietmar Maurer [Tue, 10 Dec 2013 09:46:50 +0000 (10:46 +0100)]
code cleanup

Use new helper methods.

10 years agouse new helper PVE::Tools::next_spice_port()
Dietmar Maurer [Mon, 9 Dec 2013 12:23:26 +0000 (13:23 +0100)]
use new helper PVE::Tools::next_spice_port()

10 years agospiceproxy: use POST instead of GET
Dietmar Maurer [Mon, 9 Dec 2013 07:17:52 +0000 (08:17 +0100)]
spiceproxy: use POST instead of GET

using GET is a security risk (no CSRF checks)

10 years agoadd changelog entry for last commit
Dietmar Maurer [Thu, 5 Dec 2013 05:36:00 +0000 (06:36 +0100)]
add changelog entry for last commit

10 years agoqemuserver : iothrottle add support for throttling burst max
Alexandre Derumier [Wed, 4 Dec 2013 14:12:36 +0000 (15:12 +0100)]
qemuserver : iothrottle add support for throttling burst max

http://git.qemu.org/?p=qemu.git;a=commit;h=3e9fab690d59ac15956c3733fe0794ce1ae4c4af

Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
10 years agoadd 'lsi53c810' to the list of scsi controllers
Dietmar Maurer [Wed, 4 Dec 2013 08:11:36 +0000 (09:11 +0100)]
add 'lsi53c810' to the list of scsi controllers