David Limbeck [Thu, 21 Feb 2019 15:18:53 +0000 (16:18 +0100)]
fix #2101: ipv6 ending in ':' not parsed as a string
Space or newline after ':' is recognized as a mapping and as a result an
ipv6 ending in ':' is not parsed as a string. The solution is to quote
the address. For consistency all other addresses (including mac) are
quoted.
Signed-off-by: David Limbeck <d.limbeck@proxmox.com>
Kamil Trzciński [Sun, 17 Feb 2019 12:39:37 +0000 (13:39 +0100)]
Use nr_hugepages from /proc/cmdline
Currently Proxmox VE always deallocates HugePagesTLB
when starting a new machine and it makes it impossible
to preconfigure kernel /proc/cmdline with persistent allocation.
This change makes deallocation to prefer defaults set by /proc/cmdline,
by parsing the cmdline and respecting hugepages= and hugepagesz=.
fix #2043: vm start: always stop existing systemd scopes
commit 3c23aa808ccc946bad92d9bc63b6f833c61d0f52 tried to fix a issue
where after a stop mode backup a scope could still linger around, but
it actually removed the wrong check. If we want to remove a
lingering, not yet cleaned up, scope we need to check if said scope
exists not if a VM process is still running. While they are corelated
the scope will always get cleaned up _after_ it's processes are gone.
Should fix #2043, but as this is seemingly not that easy to fix one
for all I'll put the should as disclaimer here.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com> Cc: Wolfgang Bumiller <w.bumiller@proxmox.com>
Rhonda D'Vine [Wed, 30 Jan 2019 13:43:38 +0000 (14:43 +0100)]
Fix #1924: add snapshot parameter
The qm CLI command offer the config and showcmd functions. Both of those
outputs may vary with respect to a given snapshot. This adds a switch
that shows the corresponding snapshot's config and command line.
The code needs a newer libpve-guest-common-perl, thus bumping the
dependency.
Dominik Csapak [Mon, 21 Jan 2019 14:56:54 +0000 (15:56 +0100)]
allow explicit set vga with gpu passthrough
this patch allows the user to explicitely set a virtual vga,
even when using the 'x-vga' flag, this is sometimes necessary,
as some users need the 'x-vga' flag on the pci device,
but still want to use a virtual vga
Thomas Lamprecht [Thu, 17 Jan 2019 09:53:36 +0000 (10:53 +0100)]
migrate: fix local disk migration with online VMs
commit 4530494bf9f3d45c4a405c53ef3688e641f6bd8e introduced an
regression with local disk migrations if the VM is online and thus
needs to live migrated and no target storage was passed as parameter.
We made the hack to write "1" to the targetstorage option in this
case obsolete, but it was still used on deciding if there are any
drives to mirror at all. Here it is enough to check if there are any
'online_local_volumes' because that hash gets only filled if we can
and are told to live mirror local disk on migrations anyway. Also,
we abort early if local disks are found and the 'with-local-disks'
option is not set.
This was reported at:
https://forum.proxmox.com/threads/livemigration-with-localdisk-doesnt-coppy-and-data-from-the-hdds-anymore.50744/
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Thomas Lamprecht [Thu, 20 Dec 2018 08:55:11 +0000 (09:55 +0100)]
fix local disk migration when no target storage is set
the check for targetstorage in:
if ($self->{running} && $self->{opts}->{targetstorage} && $local_volumes->{$volid}->{ref} eq 'config') {
was obsolete, as we always set the tragetstorage opts variable to '1'
in a broader "use same sid for remote local" check above.
So removing it leads to the same if truthtable but fixes the
check if we should fallback to the volume's SID if targetstorage is
not set, as else it seemed to be always set, and '1' is naturally not
a correct stroage ID.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com> Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Dominik Csapak [Mon, 17 Dec 2018 12:57:08 +0000 (13:57 +0100)]
add win7 pcie quirk
Win7 is very picky about pcie assignments and fails with
'error 12' the way we add hospci devices.
To combat that, we simply give the hostpci device a normal port
instead.
Start with address 0x10, so that we have space before those devices,
and between them and the ones configured in pve-q35.cfg should we
need it in the future.
Thomas Lamprecht [Mon, 10 Dec 2018 17:00:03 +0000 (18:00 +0100)]
t/cfg2cmd: ensure tests can pass again
the change to use a real diff to check tests, with it's nicer output
for failing cases, failed to ensure that the test system knew if a
case passed - earlier done indirectly by 'is_deeply', thus add a
manual 'pass' call if the tes was OK.
Further, the firs test failing aborted everything, which isn't to
nice, normally it's much more convenient if all test get run and one
can view at the results of all of them. So put the diff in an eval
and fail/note the test manually.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com> Acked-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Dominik Csapak [Thu, 6 Dec 2018 09:17:26 +0000 (10:17 +0100)]
vga: allow 'none'
so that one can explicitly disable the vga without having to specify
a serial port as display, this is mostly useful for very special
and custom gpu passthrough setups which have to be specified with
'args' and for setups which do not care about any display (not even serial)
Dominik Csapak [Thu, 6 Dec 2018 09:17:25 +0000 (10:17 +0100)]
fix #1267: move args to the end of qemu commandline
there is nothing that should be really affected by this, but
even then, this option is only for experts and people using this
should know what they are doing
Dominik Csapak [Tue, 27 Nov 2018 10:32:17 +0000 (11:32 +0100)]
fix check if machine type is q35
When live migrating, with a q35 machine will get the qemu version
encoded in the machine type, for example,'pc-q35-2.12', so we need to
allow this too and cannot expect that all q35 machine have
q35' in verbatim as their type.
So, when migrating such a machine live, we missed to include the q35
cfg because we didn't allowed versioned q35 machine types, which then
failed the migration.
Dominik Csapak [Tue, 20 Nov 2018 16:13:39 +0000 (17:13 +0100)]
add mediated devices support
with this, we are able to create and use mediated devices,
which include Intel GVT-g (aka KVMGT) and Nvidia vGPUs, and probably more
types of devices in the future
Dominik Csapak [Wed, 14 Nov 2018 13:59:56 +0000 (14:59 +0100)]
add qmeventd
this adds a program that can listen to qemu qmp events on a given socket
and if a shutdown event followed by a disconnected socket occurs,
executes qm cleanup with arguments that indicate if the
vm was closed gracefully and whether the guest initiated it
this is useful if we want to cleanup after the qemu process exited,
e.g. tap devices, vgpus, etc.
On arm we start off with a pcie bridge pcie.0. We need a
keyboard in addition to the tablet device, and we need to
connect both to an 'ehci' controller.
To do all this, we also pass the $arch variable through a
whole lot of function calls to ultimately also adapt the
hotplug code to take care of the new keyboard device.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
David Limbeck [Thu, 25 Oct 2018 09:32:45 +0000 (11:32 +0200)]
fix #1959: add fallback for 'auto' previously set by SLAAC
SLAAC previously set 'auto' which is not supported by nocloud network
config. On an up-to-date Ubuntu this should work as it uses 'dhcp' for
both dhcp and SLAAC. For others it was invalid anyway.
Signed-off-by: David Limbeck <d.limbeck@proxmox.com>
Dominik Csapak [Fri, 9 Nov 2018 12:31:09 +0000 (13:31 +0100)]
fix #1952: make vga memory configurable
we change 'vga' to a property string and add a 'memory' property
with this, the user can better control the memory given to the virtual
gpu, this is especially useful for spice/qxl since high resolutions need
more memory
Nick Chevsky [Sun, 28 Oct 2018 20:41:46 +0000 (16:41 -0400)]
Add `ssd` property to IDE, SATA, and SCSI drives
When enabled, the `ssd` property exposes drives as SSDs (rather than
rotational hard disks) by setting QEMU's `rotation_rate` property [1,
2] on `ide-hd`, `scsi-block`, and `scsi-hd` devices. This is required
to enable support for TRIM and SSD-specific optimizations in certain
guest operating systems that are limited to emulated controller types
(IDE, AHCI, and non-VirtIO SCSI).
This change also unifies the diverging IDE and SATA code paths in
QemuServer::print_drivedevice_full(), which suffered from:
* Code duplication: The only differences between IDE and SATA were in
bus-unit specification and maximum device counts.
* Inconsistent implementation: The IDE code used the new `ide-hd`
and `ide-cd` device types, whereas SATA still relied on the deprecated
`ide-drive` [3, 4] (which doesn't support `rotation_rate`).
* Different feature sets: The IDE code exposed a `model` property that
the SATA code didn't, even though QEMU supports it for both.
Instead of our own. The code is almost the same, but the
upstream implementation uses qemu's transactional system and
performs a drain() on the block device first. This seems to
help avoid some issues we run into with qcow2 files when
creating snapshots.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
this adds a VM Generation ID device uses by Windows (Server) to determine
some specific actions that may have happened with the vm
such as rollback, restore, etc.
projects / qemu-server.git / log