use PVE::Cluster qw(cfs_register_file cfs_read_file cfs_write_file);
use PVE::CGroup;
+use PVE::CpuSet;
use PVE::DataCenterConfig;
use PVE::Exception qw(raise raise_param_exc);
use PVE::Format qw(render_duration render_bytes);
use PVE::JSONSchema qw(get_standard_option parse_property_string);
use PVE::ProcFSTools;
use PVE::PBSClient;
+use PVE::RESTEnvironment qw(log_warn);
use PVE::RPCEnvironment;
use PVE::Storage;
use PVE::SysFSTools;
use PVE::QMPClient;
use PVE::QemuConfig;
-use PVE::QemuServer::Helpers qw(min_version config_aware_timeout);
+use PVE::QemuServer::Helpers qw(min_version config_aware_timeout windows_version);
use PVE::QemuServer::Cloudinit;
use PVE::QemuServer::CGroup;
use PVE::QemuServer::CPUConfig qw(print_cpu_device get_cpu_options);
my $EDK2_FW_BASE = '/usr/share/pve-edk2-firmware/';
my $OVMF = {
x86_64 => {
+ '4m-no-smm' => [
+ "$EDK2_FW_BASE/OVMF_CODE_4M.fd",
+ "$EDK2_FW_BASE/OVMF_VARS_4M.fd",
+ ],
+ '4m-no-smm-ms' => [
+ "$EDK2_FW_BASE/OVMF_CODE_4M.fd",
+ "$EDK2_FW_BASE/OVMF_VARS_4M.ms.fd",
+ ],
'4m' => [
"$EDK2_FW_BASE/OVMF_CODE_4M.secboot.fd",
"$EDK2_FW_BASE/OVMF_VARS_4M.fd",
my $cpuinfo = PVE::ProcFSTools::read_cpuinfo();
-# Note about locking: we use flock on the config file protect
-# against concurent actions.
-# Aditionaly, we have a 'lock' setting in the config file. This
-# can be set to 'migrate', 'backup', 'snapshot' or 'rollback'. Most actions are not
-# allowed when such lock is set. But you can ignore this kind of
-# lock with the --skiplock flag.
+# Note about locking: we use flock on the config file protect against concurent actions.
+# Aditionaly, we have a 'lock' setting in the config file. This can be set to 'migrate',
+# 'backup', 'snapshot' or 'rollback'. Most actions are not allowed when such lock is set.
+# But you can ignore this kind of lock with the --skiplock flag.
cfs_register_file('/qemu-server/',
\&parse_vm_config,
optional => 1,
});
-
-sub map_storage {
- my ($map, $source) = @_;
-
- return $source if !defined($map);
-
- return $map->{entries}->{$source}
- if $map->{entries} && defined($map->{entries}->{$source});
-
- return $map->{default} if $map->{default};
-
- # identity (fallback)
- return $source;
-}
-
-PVE::JSONSchema::register_standard_option('pve-targetstorage', {
- description => "Mapping from source to target storages. Providing only a single storage ID maps all source storages to that storage. Providing the special value '1' will map each source storage to itself.",
- type => 'string',
- format => 'storagepair-list',
- optional => 1,
-});
-
#no warnings 'redefine';
my $nodename_cache;
my $agent_fmt = {
enabled => {
- description => "Enable/disable Qemu GuestAgent.",
+ description => "Enable/disable communication with a Qemu Guest Agent (QGA) running in the VM.",
type => 'boolean',
default => 0,
default_key => 1,
default => 'std',
optional => 1,
default_key => 1,
- enum => [qw(cirrus qxl qxl2 qxl3 qxl4 none serial0 serial1 serial2 serial3 std virtio vmware)],
+ enum => [qw(cirrus qxl qxl2 qxl3 qxl4 none serial0 serial1 serial2 serial3 std virtio virtio-gl vmware)],
},
memory => {
description => "Sets the VGA memory (in MiB). Has no effect with serial display.",
type => 'string',
enum => ['/dev/urandom', '/dev/random', '/dev/hwrng'],
default_key => 1,
- description => "The file on the host to gather entropy from. In most"
- . " cases /dev/urandom should be preferred over /dev/random"
- . " to avoid entropy-starvation issues on the host. Using"
- . " urandom does *not* decrease security in any meaningful"
- . " way, as it's still seeded from real entropy, and the"
- . " bytes provided will most likely be mixed with real"
- . " entropy on the guest as well. /dev/hwrng can be used"
- . " to pass through a hardware RNG from the host.",
+ description => "The file on the host to gather entropy from. In most cases '/dev/urandom'"
+ ." should be preferred over '/dev/random' to avoid entropy-starvation issues on the"
+ ." host. Using urandom does *not* decrease security in any meaningful way, as it's"
+ ." still seeded from real entropy, and the bytes provided will most likely be mixed"
+ ." with real entropy on the guest as well. '/dev/hwrng' can be used to pass through"
+ ." a hardware RNG from the host.",
},
max_bytes => {
type => 'integer',
- description => "Maximum bytes of entropy injected into the guest every"
- . " 'period' milliseconds. Prefer a lower value when using"
- . " /dev/random as source. Use 0 to disable limiting"
- . " (potentially dangerous!).",
+ description => "Maximum bytes of entropy allowed to get injected into the guest every"
+ ." 'period' milliseconds. Prefer a lower value when using '/dev/random' as source. Use"
+ ." `0` to disable limiting (potentially dangerous!).",
optional => 1,
- # default is 1 KiB/s, provides enough entropy to the guest to avoid
- # boot-starvation issues (e.g. systemd etc...) while allowing no chance
- # of overwhelming the host, provided we're reading from /dev/urandom
+ # default is 1 KiB/s, provides enough entropy to the guest to avoid boot-starvation issues
+ # (e.g. systemd etc...) while allowing no chance of overwhelming the host, provided we're
+ # reading from /dev/urandom
default => 1024,
},
period => {
type => 'integer',
- description => "Every 'period' milliseconds the entropy-injection quota"
- . " is reset, allowing the guest to retrieve another"
- . " 'max_bytes' of entropy.",
+ description => "Every 'period' milliseconds the entropy-injection quota is reset, allowing"
+ ." the guest to retrieve another 'max_bytes' of entropy.",
optional => 1,
default => 1000,
},
};
+my $meta_info_fmt = {
+ 'ctime' => {
+ type => 'integer',
+ description => "The guest creation timestamp as UNIX epoch time",
+ minimum => 0,
+ optional => 1,
+ },
+ 'creation-qemu' => {
+ type => 'string',
+ description => "The QEMU (machine) version from the time this VM was created.",
+ pattern => '\d+(\.\d+)+',
+ optional => 1,
+ },
+};
+
my $confdesc = {
onboot => {
optional => 1,
default => 0,
},
hotplug => {
- optional => 1,
- type => 'string', format => 'pve-hotplug-features',
- description => "Selectively enable hotplug features. This is a comma separated list of hotplug features: 'network', 'disk', 'cpu', 'memory' and 'usb'. Use '0' to disable hotplug completely. Value '1' is an alias for the default 'network,disk,usb'.",
+ optional => 1,
+ type => 'string', format => 'pve-hotplug-features',
+ description => "Selectively enable hotplug features. This is a comma separated list of"
+ ." hotplug features: 'network', 'disk', 'cpu', 'memory', 'usb' and 'cloudinit'. Use '0' to disable"
+ ." hotplug completely. Using '1' as value is an alias for the default `network,disk,usb`.",
default => 'network,disk,usb',
},
reboot => {
optional => 1,
type => 'number',
description => "Limit of CPU usage.",
- verbose_description => "Limit of CPU usage.\n\nNOTE: If the computer has 2 CPUs, it has total of '2' CPU time. Value '0' indicates no CPU limit.",
+ verbose_description => "Limit of CPU usage.\n\nNOTE: If the computer has 2 CPUs, it has"
+ ." total of '2' CPU time. Value '0' indicates no CPU limit.",
minimum => 0,
maximum => 128,
- default => 0,
+ default => 0,
},
cpuunits => {
optional => 1,
type => 'integer',
- description => "CPU weight for a VM, will be clamped to [1, 10000] in cgroup v2.",
+ description => "CPU weight for a VM, will be clamped to [1, 10000] in cgroup v2.",
verbose_description => "CPU weight for a VM. Argument is used in the kernel fair scheduler."
." The larger the number is, the more CPU time this VM gets. Number is relative to"
." weights of all the other running VMs.",
- minimum => 2,
+ minimum => 1,
maximum => 262144,
default => 'cgroup v1: 1024, cgroup v2: 100',
},
memory => {
optional => 1,
type => 'integer',
- description => "Amount of RAM for the VM in MB. This is the maximum available memory when you use the balloon device.",
+ description => "Amount of RAM for the VM in MB. This is the maximum available memory when"
+ ." you use the balloon device.",
minimum => 16,
default => 512,
},
balloon => {
- optional => 1,
- type => 'integer',
- description => "Amount of target RAM for the VM in MB. Using zero disables the ballon driver.",
+ optional => 1,
+ type => 'integer',
+ description => "Amount of target RAM for the VM in MB. Using zero disables the ballon driver.",
minimum => 0,
},
shares => {
- optional => 1,
- type => 'integer',
- description => "Amount of memory shares for auto-ballooning. The larger the number is, the more memory this VM gets. Number is relative to weights of all other running VMs. Using zero disables auto-ballooning. Auto-ballooning is done by pvestatd.",
+ optional => 1,
+ type => 'integer',
+ description => "Amount of memory shares for auto-ballooning. The larger the number is, the"
+ ." more memory this VM gets. Number is relative to weights of all other running VMs."
+ ." Using zero disables auto-ballooning. Auto-ballooning is done by pvestatd.",
minimum => 0,
maximum => 50000,
default => 1000,
keyboard => {
optional => 1,
type => 'string',
- description => "Keybord layout for vnc server. Default is read from the '/etc/pve/datacenter.cfg' configuration file.".
- "It should not be necessary to set it.",
+ description => "Keyboard layout for VNC server. This option is generally not required and"
+ ." is often better handled from within the guest OS.",
enum => PVE::Tools::kvmkeymaplist(),
default => undef,
},
ostype => {
optional => 1,
type => 'string',
- enum => [qw(other wxp w2k w2k3 w2k8 wvista win7 win8 win10 win11 l24 l26 solaris)],
+ enum => [qw(other wxp w2k w2k3 w2k8 wvista win7 win8 win10 win11 l24 l26 solaris)],
description => "Specify guest operating system.",
verbose_description => <<EODESC,
Specify guest operating system. This is used to enable special
boot => {
optional => 1,
type => 'string', format => 'pve-qm-boot',
- description => "Specify guest boot order. Use with 'order=', usage with"
- . " no key or 'legacy=' is deprecated.",
+ description => "Specify guest boot order. Use the 'order=' sub-property as usage with no"
+ ." key or 'legacy=' is deprecated.",
},
bootdisk => {
optional => 1,
},
agent => {
optional => 1,
- description => "Enable/disable Qemu GuestAgent and its properties.",
+ description => "Enable/disable communication with the Qemu Guest Agent and its properties.",
type => 'string',
format => $agent_fmt,
},
localtime => {
optional => 1,
type => 'boolean',
- description => "Set the real time clock to local time. This is enabled by default if ostype"
- ." indicates a Microsoft OS.",
+ description => "Set the real time clock (RTC) to local time. This is enabled by default if"
+ ." the `ostype` indicates a Microsoft Windows OS.",
},
freeze => {
optional => 1,
description => "Configure a VirtIO-based Random Number Generator.",
optional => 1,
},
+ meta => {
+ type => 'string',
+ format => $meta_info_fmt,
+ description => "Some (read-only) meta-information about this guest.",
+ optional => 1,
+ },
+ affinity => {
+ type => 'string', format => 'pve-cpuset',
+ description => "List of host cores used to execute guest processes.",
+ optional => 1,
+ },
};
my $cicustom_fmt = {
format => 'pve-volume-id',
format_description => 'volume',
},
+ vendor => {
+ type => 'string',
+ optional => 1,
+ description => 'Specify a custom file containing all vendor data passed to the VM via'
+ .' cloud-init.',
+ format => 'pve-volume-id',
+ format_description => 'volume',
+ },
};
PVE::JSONSchema::register_format('pve-qm-cicustom', $cicustom_fmt);
searchdomain => {
optional => 1,
type => 'string',
- description => "cloud-init: Sets DNS search domains for a container. Create will'
+ description => 'cloud-init: Sets DNS search domains for a container. Create will'
.' automatically use the setting from the host if neither searchdomain nor nameserver'
- .' are set.",
+ .' are set.',
},
nameserver => {
optional => 1,
type => 'string', format => 'address-list',
- description => "cloud-init: Sets DNS server IP address for a container. Create will'
+ description => 'cloud-init: Sets DNS server IP address for a container. Create will'
.' automatically use the setting from the host if neither searchdomain nor nameserver'
- .' are set.",
+ .' are set.',
},
sshkeys => {
optional => 1,
default_key => 1,
},
(map { $_ => { keyAlias => 'model', alias => 'macaddr' }} @$nic_model_list),
- bridge => {
- type => 'string',
+ bridge => get_standard_option('pve-bridge-id', {
description => $net_fmt_bridge_descr,
- format_description => 'bridge',
- pattern => '[-_.\w\d]+',
optional => 1,
- },
+ }),
queues => {
type => 'integer',
minimum => 0, maximum => 16,
$confdesc->{$key} = $confdesc_cloudinit->{$key};
}
+PVE::JSONSchema::register_format('pve-cpuset', \&pve_verify_cpuset);
+sub pve_verify_cpuset {
+ my ($set_text, $noerr) = @_;
+
+ my ($count, $members) = eval { PVE::CpuSet::parse_cpuset($set_text) };
+
+ if ($@) {
+ return if $noerr;
+ die "unable to parse cpuset option\n";
+ }
+
+ return PVE::CpuSet->new($members)->short_string();
+}
+
PVE::JSONSchema::register_format('pve-volume-id-or-qm-path', \&verify_volume_id_or_qm_path);
sub verify_volume_id_or_qm_path {
my ($volid, $noerr) = @_;
- if ($volid eq 'none' || $volid eq 'cdrom' || $volid =~ m|^/|) {
- return $volid;
- }
+ return $volid if $volid eq 'none' || $volid eq 'cdrom';
+
+ return verify_volume_id_or_absolute_path($volid, $noerr);
+}
+
+PVE::JSONSchema::register_format('pve-volume-id-or-absolute-path', \&verify_volume_id_or_absolute_path);
+sub verify_volume_id_or_absolute_path {
+ my ($volid, $noerr) = @_;
+
+ return $volid if $volid =~ m|^/|;
- # if its neither 'none' nor 'cdrom' nor a path, check if its a volume-id
$volid = eval { PVE::JSONSchema::check_format('pve-volume-id', $volid, '') };
if ($@) {
return if $noerr;
usb3 => {
optional => 1,
type => 'boolean',
- description => "Specifies whether if given host option is a USB3 device or port.",
+ description => "Specifies whether if given host option is a USB3 device or port."
+ ." For modern guests (machine version >= 7.1 and ostype l26 and windows > 7), this flag"
+ ." is irrelevant (all devices are plugged into a xhci controller).",
default => 0,
},
};
my sub extract_version {
my ($machine_type, $version) = @_;
$version = kvm_user_version() if !defined($version);
- PVE::QemuServer::Machine::extract_version($machine_type, $version)
+ return PVE::QemuServer::Machine::extract_version($machine_type, $version)
}
sub kernel_has_vhost_net {
$data = $confdesc->{hotplug}->{default} if $data eq '1';
foreach my $feature (PVE::Tools::split_list($data)) {
- if ($feature =~ m/^(network|disk|cpu|memory|usb)$/) {
+ if ($feature =~ m/^(network|disk|cpu|memory|usb|cloudinit)$/) {
$res->{$1} = 1;
} else {
die "invalid hotplug feature '$feature'\n";
# see /usr/include/scsi/sg.h
my $sg_io_hdr_t = "i i C C s I P P P I I i P C C C C S S i I I";
- my $packet = pack($sg_io_hdr_t, ord('S'), -3, length($cmd),
- length($sensebuf), 0, length($buf), $buf,
- $cmd, $sensebuf, 6000);
+ my $packet = pack(
+ $sg_io_hdr_t, ord('S'), -3, length($cmd), length($sensebuf), 0, length($buf), $buf, $cmd, $sensebuf, 6000
+ );
$ret = ioctl($fh, $SG_IO, $packet);
if (!$ret) {
}
my $res = {};
- (my $byte0, my $byte1, $res->{vendor},
- $res->{product}, $res->{revision}) = unpack("C C x6 A8 A16 A4", $buf);
+ $res->@{qw(type removable vendor product revision)} = unpack("C C x6 A8 A16 A4", $buf);
- $res->{removable} = $byte1 & 128 ? 1 : 0;
- $res->{type} = $byte0 & 31;
+ $res->{removable} = $res->{removable} & 128 ? 1 : 0;
+ $res->{type} &= 0x1F;
return $res;
}
# we use uhci for old VMs because tablet driver was buggy in older qemu
my $usbbus;
- if (PVE::QemuServer::Machine::machine_type_is_q35($conf) || $arch eq 'aarch64') {
+ if ($q35 || $arch eq 'aarch64') {
$usbbus = 'ehci';
} else {
$usbbus = 'uhci';
}
sub print_keyboarddevice_full {
- my ($conf, $arch, $machine) = @_;
+ my ($conf, $arch) = @_;
return if $arch ne 'aarch64';
$opts .= ",snapshot=$v";
}
+ if (defined($drive->{ro})) { # ro maps to QEMUs `readonly`, which accepts `on` or `off` only
+ $opts .= ",readonly=" . ($drive->{ro} ? 'on' : 'off');
+ }
+
foreach my $type (['', '-total'], [_rd => '-read'], [_wr => '-write']) {
my ($dir, $qmpname) = @$type;
if (my $v = $drive->{"mbps$dir"}) {
# sometimes, just plain disable...
my $lvm_no_io_uring = $scfg && $scfg->{type} eq 'lvm';
+ # io_uring causes problems when used with CIFS since kernel 5.15
+ # Some discussion: https://www.spinics.net/lists/linux-cifs/msg26734.html
+ my $cifs_no_io_uring = $scfg && $scfg->{type} eq 'cifs';
+
if (!$drive->{aio}) {
- if ($io_uring && !$rbd_no_io_uring && !$lvm_no_io_uring) {
+ if ($io_uring && !$rbd_no_io_uring && !$lvm_no_io_uring && !$cifs_no_io_uring) {
# io_uring supports all cache modes
$opts .= ",aio=io_uring";
} else {
my ($pbs_conf, $pbs_name) = @_;
my $blockdev = "driver=pbs,node-name=$pbs_name,read-only=on";
$blockdev .= ",repository=$pbs_conf->{repository}";
+ $blockdev .= ",namespace=$pbs_conf->{namespace}" if $pbs_conf->{namespace};
$blockdev .= ",snapshot=$pbs_conf->{snapshot}";
$blockdev .= ",archive=$pbs_conf->{archive}";
$blockdev .= ",keyfile=$pbs_conf->{keyfile}" if $pbs_conf->{keyfile};
'std' => 'VGA',
'vmware' => 'vmware-svga',
'virtio' => 'virtio-vga',
+ 'virtio-gl' => 'virtio-vga-gl',
};
sub print_vga_device {
my $memory = "";
if ($vgamem_mb) {
- if ($vga->{type} eq 'virtio') {
+ if ($vga->{type} =~ /^virtio/) {
my $bytes = PVE::Tools::convert_size($vgamem_mb, "mb" => "b");
$memory = ",max_hostmem=$bytes";
} elsif ($qxlnum) {
my $q35 = PVE::QemuServer::Machine::machine_type_is_q35($conf);
my $vgaid = "vga" . ($id // '');
my $pciaddr;
-
if ($q35 && $vgaid eq 'vga') {
# the first display uses pcie.0 bus on q35 machines
$pciaddr = print_pcie_addr($vgaid, $bridges, $arch, $machine);
$pciaddr = print_pci_addr($vgaid, $bridges, $arch, $machine);
}
+ if ($vga->{type} eq 'virtio-gl') {
+ my $base = '/usr/lib/x86_64-linux-gnu/lib';
+ die "missing libraries for '$vga->{type}' detected! Please install 'libgl1' and 'libegl1'\n"
+ if !-e "${base}EGL.so.1" || !-e "${base}GL.so.1";
+
+ die "no DRM render node detected (/dev/dri/renderD*), no GPU? - needed for '$vga->{type}' display\n"
+ if !PVE::Tools::dir_glob_regex('/dev/dri/', "renderD.*");
+ }
+
return "$type,id=${vgaid}${memory}${max_outputs}${pciaddr}${edidoff}";
}
if (drive_is_cloudinit($drive)) {
eval { PVE::Storage::vdisk_free($storecfg, $drive->{file}) };
warn $@ if $@;
+ delete $conf->{cloudinit};
} elsif (!drive_is_cdrom($drive)) {
my $volid = $drive->{file};
if (vm_is_volid_owner($storecfg, $vmid, $volid)) {
return $res;
}
+sub parse_meta_info {
+ my ($value) = @_;
+
+ return if !$value;
+
+ my $res = eval { parse_property_string($meta_info_fmt, $value) };
+ warn $@ if $@;
+ return $res;
+}
+
+sub new_meta_info_string {
+ my () = @_; # for now do not allow to override any value
+
+ return PVE::JSONSchema::print_property_string(
+ {
+ 'creation-qemu' => kvm_user_version(),
+ ctime => "". int(time()),
+ },
+ $meta_info_fmt
+ );
+}
+
+sub qemu_created_version_fixups {
+ my ($conf, $forcemachine, $kvmver) = @_;
+
+ my $meta = parse_meta_info($conf->{meta}) // {};
+ my $forced_vers = PVE::QemuServer::Machine::extract_version($forcemachine);
+
+ # check if we need to apply some handling for VMs that always use the latest machine version but
+ # had a machine version transition happen that affected HW such that, e.g., an OS config change
+ # would be required (we do not want to pin machine version for non-windows OS type)
+ if (
+ (!defined($conf->{machine}) || $conf->{machine} =~ m/^(?:pc|q35|virt)$/) # non-versioned machine
+ && (!defined($meta->{'creation-qemu'}) || !min_version($meta->{'creation-qemu'}, 6, 1)) # created before 6.1
+ && (!$forced_vers || min_version($forced_vers, 6, 1)) # handle snapshot-rollback/migrations
+ && min_version($kvmver, 6, 1) # only need to apply the change since 6.1
+ ) {
+ my $q35 = PVE::QemuServer::Machine::machine_type_is_q35($conf);
+ if ($q35 && $conf->{ostype} && $conf->{ostype} eq 'l26') {
+ # this changed to default-on in Q 6.1 for q35 machines, it will mess with PCI slot view
+ # and thus with the predictable interface naming of systemd
+ return ['-global', 'ICH9-LPC.acpi-pci-hotplug-with-bridge-support=off'];
+ }
+ }
+ return;
+}
+
PVE::JSONSchema::register_format('pve-qm-usb-device', \&verify_usb_device);
sub verify_usb_device {
my ($value, $noerr) = @_;
# add JSON properties for create and set function
sub json_config_properties {
- my $prop = shift;
+ my ($prop, $with_disk_alloc) = @_;
+
+ my $skip_json_config_opts = {
+ parent => 1,
+ snaptime => 1,
+ vmstate => 1,
+ runningmachine => 1,
+ runningcpu => 1,
+ meta => 1,
+ };
foreach my $opt (keys %$confdesc) {
- next if $opt eq 'parent' || $opt eq 'snaptime' || $opt eq 'vmstate' ||
- $opt eq 'runningmachine' || $opt eq 'runningcpu';
- $prop->{$opt} = $confdesc->{$opt};
+ next if $skip_json_config_opts->{$opt};
+
+ if ($with_disk_alloc && is_valid_drivename($opt)) {
+ $prop->{$opt} = $PVE::QemuServer::Drive::drivedesc_hash_with_alloc->{$opt};
+ } else {
+ $prop->{$opt} = $confdesc->{$opt};
+ }
}
return $prop;
}
+# Properties that we can read from an OVF file
+sub json_ovf_properties {
+ my $prop = {};
+
+ for my $device (PVE::QemuServer::Drive::valid_drive_names()) {
+ $prop->{$device} = {
+ type => 'string',
+ format => 'pve-volume-id-or-absolute-path',
+ description => "Disk image that gets imported to $device",
+ optional => 1,
+ };
+ }
+
+ $prop->{cores} = {
+ type => 'integer',
+ description => "The number of CPU cores.",
+ optional => 1,
+ };
+ $prop->{memory} = {
+ type => 'integer',
+ description => "Amount of RAM for the VM in MB.",
+ optional => 1,
+ };
+ $prop->{name} = {
+ type => 'string',
+ description => "Name of the VM.",
+ optional => 1,
+ };
+
+ return $prop;
+}
+
# return copy of $confdesc_cloudinit to generate documentation
sub cloudinit_config_properties {
}
sub parse_vm_config {
- my ($filename, $raw) = @_;
+ my ($filename, $raw, $strict) = @_;
return if !defined($raw);
digest => Digest::SHA::sha1_hex($raw),
snapshots => {},
pending => {},
+ cloudinit => {},
+ };
+
+ my $handle_error = sub {
+ my ($msg) = @_;
+
+ if ($strict) {
+ die $msg;
+ } else {
+ warn $msg;
+ }
};
$filename =~ m|/qemu-server/(\d+)\.conf$|
$descr = undef;
$conf = $res->{$section} = {};
next;
+ } elsif ($line =~ m/^\[special:cloudinit\]\s*$/i) {
+ $section = 'cloudinit';
+ $descr = undef;
+ $conf = $res->{$section} = {};
+ next;
} elsif ($line =~ m/^\[([a-z][a-z0-9_\-]+)\]\s*$/i) {
$section = $1;
next;
}
- if ($line =~ m/^\#(.*)\s*$/) {
+ if ($line =~ m/^\#(.*)$/) {
$descr = '' if !defined($descr);
$descr .= PVE::Tools::decode_text($1) . "\n";
next;
if ($section eq 'pending') {
$conf->{delete} = $value; # we parse this later
} else {
- warn "vm $vmid - propertry 'delete' is only allowed in [PENDING]\n";
+ $handle_error->("vm $vmid - property 'delete' is only allowed in [PENDING]\n");
}
} elsif ($line =~ m/^([a-z][a-z_]*\d*):\s*(.+?)\s*$/) {
my $key = $1;
my $value = $2;
eval { $value = check_type($key, $value); };
if ($@) {
- warn "vm $vmid - unable to parse value of '$key' - $@";
+ $handle_error->("vm $vmid - unable to parse value of '$key' - $@");
} else {
$key = 'ide2' if $key eq 'cdrom';
my $fmt = $confdesc->{$key}->{format};
$v->{file} = $volid;
$value = print_drive($v);
} else {
- warn "vm $vmid - unable to parse value of '$key'\n";
+ $handle_error->("vm $vmid - unable to parse value of '$key'\n");
next;
}
}
$conf->{$key} = $value;
}
} else {
- warn "vm $vmid - unable to parse config: $line\n";
+ $handle_error->("vm $vmid - unable to parse config: $line\n");
}
}
foreach my $key (keys %$cref) {
next if $key eq 'digest' || $key eq 'description' || $key eq 'snapshots' ||
- $key eq 'snapstate' || $key eq 'pending';
+ $key eq 'snapstate' || $key eq 'pending' || $key eq 'cloudinit';
my $value = $cref->{$key};
if ($key eq 'delete') {
die "propertry 'delete' is only allowed in [PENDING]\n"
&$cleanup_config($conf->{pending}, 1);
+ &$cleanup_config($conf->{cloudinit});
+
foreach my $snapname (keys %{$conf->{snapshots}}) {
die "internal error: snapshot name '$snapname' is forbidden" if lc($snapname) eq 'pending';
&$cleanup_config($conf->{snapshots}->{$snapname}, undef, $snapname);
}
foreach my $key (sort keys %$conf) {
- next if $key =~ /^(digest|description|pending|snapshots)$/;
+ next if $key =~ /^(digest|description|pending|cloudinit|snapshots)$/;
$raw .= "$key: $conf->{$key}\n";
}
return $raw;
$raw .= &$generate_raw_config($conf->{pending}, 1);
}
+ if (scalar(keys %{$conf->{cloudinit}})){
+ $raw .= "\n[special:cloudinit]\n";
+ $raw .= &$generate_raw_config($conf->{cloudinit});
+ }
+
foreach my $snapname (sort keys %{$conf->{snapshots}}) {
$raw .= "\n[$snapname]\n";
$raw .= &$generate_raw_config($conf->{snapshots}->{$snapname});
return $machine;
}
-sub get_ovmf_files($$) {
- my ($arch, $efidisk) = @_;
+sub get_ovmf_files($$$) {
+ my ($arch, $efidisk, $smm) = @_;
my $types = $OVMF->{$arch}
or die "no OVMF images known for architecture '$arch'\n";
my $type = 'default';
if (defined($efidisk->{efitype}) && $efidisk->{efitype} eq '4m') {
- $type = $efidisk->{'pre-enrolled-keys'} ? "4m-ms" : "4m";
+ $type = $smm ? "4m" : "4m-no-smm";
+ $type .= '-ms' if $efidisk->{'pre-enrolled-keys'};
}
return $types->{$type}->@*;
return \@flags;
}
-my sub get_cpuunits {
- my ($conf) = @_;
- return $conf->{cpuunits} // (PVE::CGroup::cgroup_mode() == 2 ? 100 : 1024);
+# Since commit 277d33454f77ec1d1e0bc04e37621e4dd2424b67 in pve-qemu, smm is not off by default
+# anymore. But smm=off seems to be required when using SeaBIOS and serial display.
+my sub should_disable_smm {
+ my ($conf, $vga) = @_;
+
+ return (!defined($conf->{bios}) || $conf->{bios} eq 'seabios') &&
+ $vga->{type} && $vga->{type} =~ m/^(serial\d+|none)$/;
}
+
sub config_to_command {
my ($storecfg, $vmid, $conf, $defaults, $forcemachine, $forcecpu,
$pbs_backing) = @_;
my $cmd = [];
- my $globalFlags = [];
- my $machineFlags = [];
- my $rtcFlags = [];
+ my ($globalFlags, $machineFlags, $rtcFlags) = ([], [], []);
my $devices = [];
- my $pciaddr = '';
my $bridges = {};
my $ostype = $conf->{ostype};
my $winversion = windows_version($ostype);
my $use_old_bios_files = undef;
($use_old_bios_files, $machine_type) = qemu_use_old_bios_files($machine_type);
- my $cpuunits = get_cpuunits($conf);
+ if ($conf->{affinity}) {
+ push @$cmd, "/usr/bin/taskset";
+ push @$cmd, "--cpu-list";
+ push @$cmd, "--all-tasks";
+ push @$cmd, $conf->{affinity};
+ }
push @$cmd, $kvm_binary;
my $vmname = $conf->{name} || "vm$vmid";
- push @$cmd, '-name', $vmname;
+ push @$cmd, '-name', "$vmname,debug-threads=on";
push @$cmd, '-no-shutdown';
$d = parse_drive('efidisk0', $efidisk);
}
- my ($ovmf_code, $ovmf_vars) = get_ovmf_files($arch, $d);
+ my ($ovmf_code, $ovmf_vars) = get_ovmf_files($arch, $d, $q35);
die "uefi base image '$ovmf_code' not found\n" if ! -f $ovmf_code;
my ($path, $format);
$read_only_str = ',readonly=on' if drive_is_read_only($conf, $d);
} else {
- warn "no efidisk configured! Using temporary efivars disk.\n";
+ log_warn("no efidisk configured! Using temporary efivars disk.");
$path = "/tmp/$vmid-ovmf.fd";
PVE::Tools::file_copy($ovmf_vars, $path, -s $ovmf_vars);
$format = 'raw';
push @$cmd, '-drive', "if=pflash,unit=1$cache,format=$format,id=drive-efidisk0$size_str,file=${path}${read_only_str}";
}
- # load q35 config
- if ($q35) {
+ if ($q35) { # tell QEMU to load q35 config early
# we use different pcie-port hardware for qemu >= 4.0 for passthrough
if (min_version($machine_version, 4, 0)) {
push @$devices, '-readconfig', '/usr/share/qemu-server/pve-q35-4.0.cfg';
}
}
+ if (defined(my $fixups = qemu_created_version_fixups($conf, $forcemachine, $kvmver))) {
+ push @$cmd, $fixups->@*;
+ }
+
if ($conf->{vmgenid}) {
push @$devices, '-device', 'vmgenid,guid='.$conf->{vmgenid};
}
# add usb controllers
my @usbcontrollers = PVE::QemuServer::USB::get_usb_controllers(
- $conf, $bridges, $arch, $machine_type, $usbdesc->{format}, $MAX_USB_DEVICES);
+ $conf, $bridges, $arch, $machine_type, $usbdesc->{format}, $MAX_USB_DEVICES, $machine_version);
push @$devices, @usbcontrollers if @usbcontrollers;
my $vga = parse_vga($conf->{vga});
}
# enable absolute mouse coordinates (needed by vnc)
- my $tablet;
- if (defined($conf->{tablet})) {
- $tablet = $conf->{tablet};
- } else {
+ my $tablet = $conf->{tablet};
+ if (!defined($tablet)) {
$tablet = $defaults->{tablet};
$tablet = 0 if $qxlnum; # disable for spice because it is not needed
$tablet = 0 if $vga->{type} =~ m/^serial\d+$/; # disable if we use serial terminal (no vga card)
$usb_dev_features->{spice_usb3} = 1 if min_version($machine_version, 4, 0);
my @usbdevices = PVE::QemuServer::USB::get_usb_devices(
- $conf, $usbdesc->{format}, $MAX_USB_DEVICES, $usb_dev_features, $bootorder);
+ $conf, $usbdesc->{format}, $MAX_USB_DEVICES, $usb_dev_features, $bootorder, $machine_version);
push @$devices, @usbdevices if @usbdevices;
# serial devices
for (my $i = 0; $i < $MAX_SERIAL_PORTS; $i++) {
- if (my $path = $conf->{"serial$i"}) {
- if ($path eq 'socket') {
- my $socket = "/var/run/qemu-server/${vmid}.serial$i";
- push @$devices, '-chardev', "socket,id=serial$i,path=$socket,server=on,wait=off";
- # On aarch64, serial0 is the UART device. Qemu only allows
- # connecting UART devices via the '-serial' command line, as
- # the device has a fixed slot on the hardware...
- if ($arch eq 'aarch64' && $i == 0) {
- push @$devices, '-serial', "chardev:serial$i";
- } else {
- push @$devices, '-device', "isa-serial,chardev=serial$i";
- }
+ my $path = $conf->{"serial$i"} or next;
+ if ($path eq 'socket') {
+ my $socket = "/var/run/qemu-server/${vmid}.serial$i";
+ push @$devices, '-chardev', "socket,id=serial$i,path=$socket,server=on,wait=off";
+ # On aarch64, serial0 is the UART device. Qemu only allows
+ # connecting UART devices via the '-serial' command line, as
+ # the device has a fixed slot on the hardware...
+ if ($arch eq 'aarch64' && $i == 0) {
+ push @$devices, '-serial', "chardev:serial$i";
} else {
- die "no such serial device\n" if ! -c $path;
- push @$devices, '-chardev', "tty,id=serial$i,path=$path";
push @$devices, '-device', "isa-serial,chardev=serial$i";
}
+ } else {
+ die "no such serial device\n" if ! -c $path;
+ push @$devices, '-chardev', "tty,id=serial$i,path=$path";
+ push @$devices, '-device', "isa-serial,chardev=serial$i";
}
}
my $allowed_vcpus = $cpuinfo->{cpus};
- die "MAX $allowed_vcpus vcpus allowed per VM on this node\n"
- if ($allowed_vcpus < $maxcpus);
-
- if($hotplug_features->{cpu} && min_version($machine_version, 2, 7)) {
+ die "MAX $allowed_vcpus vcpus allowed per VM on this node\n" if ($allowed_vcpus < $maxcpus);
+ if ($hotplug_features->{cpu} && min_version($machine_version, 2, 7)) {
push @$cmd, '-smp', "1,sockets=$sockets,cores=$cores,maxcpus=$maxcpus";
for (my $i = 2; $i <= $vcpus; $i++) {
my $cpustr = print_cpu_device($conf,$i);
if ($vga->{type} && $vga->{type} !~ m/^serial\d+$/ && $vga->{type} ne 'none'){
push @$devices, '-device', print_vga_device(
$conf, $vga, $arch, $machine_version, $machine_type, undef, $qxlnum, $bridges);
+
+ push @$cmd, '-display', 'egl-headless,gl=core' if $vga->{type} eq 'virtio-gl'; # VIRGL
+
my $socket = PVE::QemuServer::Helpers::vnc_socket($vmid);
push @$cmd, '-vnc', "unix:$socket,password=on";
} else {
my $spice_port;
- if ($qxlnum) {
+ if ($qxlnum || $vga->{type} =~ /^virtio/) {
if ($qxlnum > 1) {
if ($winversion){
for (my $i = 1; $i < $qxlnum; $i++){
# enable balloon by default, unless explicitly disabled
if (!defined($conf->{balloon}) || $conf->{balloon}) {
- $pciaddr = print_pci_addr("balloon0", $bridges, $arch, $machine_type);
- push @$devices, '-device', "virtio-balloon-pci,id=balloon0$pciaddr";
+ my $pciaddr = print_pci_addr("balloon0", $bridges, $arch, $machine_type);
+ my $ballooncmd = "virtio-balloon-pci,id=balloon0$pciaddr";
+ $ballooncmd .= ",free-page-reporting=on" if min_version($machine_version, 6, 2);
+ push @$devices, '-device', $ballooncmd;
}
if ($conf->{watchdog}) {
my $wdopts = parse_watchdog($conf->{watchdog});
- $pciaddr = print_pci_addr("watchdog", $bridges, $arch, $machine_type);
+ my $pciaddr = print_pci_addr("watchdog", $bridges, $arch, $machine_type);
my $watchdog = $wdopts->{model} || 'i6300esb';
push @$devices, '-device', "$watchdog$pciaddr";
push @$devices, '-watchdog-action', $wdopts->{action} if $wdopts->{action};
die "scsi$drive->{index}: machine version 4.1~pve2 or higher is required to use more than 14 SCSI disks\n"
if $drive->{index} > 13 && !&$version_guard(4, 1, 2);
- $pciaddr = print_pci_addr("$controller_prefix$controller", $bridges, $arch, $machine_type);
+ my $pciaddr = print_pci_addr("$controller_prefix$controller", $bridges, $arch, $machine_type);
my $scsihw_type = $scsihw =~ m/^virtio-scsi-single/ ? "virtio-scsi-pci" : $scsihw;
my $iothread = '';
$iothread .= ",iothread=iothread-$controller_prefix$controller";
push @$cmd, '-object', "iothread,id=iothread-$controller_prefix$controller";
} elsif ($drive->{iothread}) {
- warn "iothread is only valid with virtio disk or virtio-scsi-single controller, ignoring\n";
+ log_warn(
+ "iothread is only valid with virtio disk or virtio-scsi-single controller, ignoring\n"
+ );
}
my $queues = '';
if ($drive->{interface} eq 'sata') {
my $controller = int($drive->{index} / $PVE::QemuServer::Drive::MAX_SATA_DISKS);
- $pciaddr = print_pci_addr("ahci$controller", $bridges, $arch, $machine_type);
+ my $pciaddr = print_pci_addr("ahci$controller", $bridges, $arch, $machine_type);
push @$devices, '-device', "ahci,id=ahci$controller,multifunction=on$pciaddr"
if !$ahcicontroller->{$controller};
$ahcicontroller->{$controller}=1;
# pci.4 is nested in pci.1
$bridges->{1} = 1 if $bridges->{4};
- if (!$q35) {
- # add pci bridges
- if (min_version($machine_version, 2, 3)) {
+ if (!$q35) { # add pci bridges
+ if (min_version($machine_version, 2, 3)) {
$bridges->{1} = 1;
$bridges->{2} = 1;
}
-
$bridges->{3} = 1 if $scsihw =~ m/^virtio-scsi-single/;
-
}
for my $k (sort {$b cmp $a} keys %$bridges) {
if ($k == 2 && $legacy_igd) {
$k_name = "$k-igd";
}
- $pciaddr = print_pci_addr("pci.$k_name", undef, $arch, $machine_type);
-
+ my $pciaddr = print_pci_addr("pci.$k_name", undef, $arch, $machine_type);
my $devstr = "pci-bridge,id=pci.$k,chassis_nr=$k$pciaddr";
- if ($q35) {
- # add after -readconfig pve-q35.cfg
+
+ if ($q35) { # add after -readconfig pve-q35.cfg
splice @$devices, 2, 0, '-device', $devstr;
} else {
unshift @$devices, '-device', $devstr if $k > 0;
push @$machineFlags, 'accel=tcg';
}
+ push @$machineFlags, 'smm=off' if should_disable_smm($conf, $vga);
+
my $machine_type_min = $machine_type;
if ($add_pve_version) {
$machine_type_min =~ s/\+pve\d+$//;
qemu_add_pci_bridge($storecfg, $conf, $vmid, $deviceid, $arch, $machine_type);
if ($deviceid eq 'tablet') {
-
qemu_deviceadd($vmid, print_tabletdevice_full($conf, $arch));
-
} elsif ($deviceid eq 'keyboard') {
-
qemu_deviceadd($vmid, print_keyboarddevice_full($conf, $arch));
-
} elsif ($deviceid =~ m/^usb(\d+)$/) {
-
die "usb hotplug currently not reliable\n";
# since we can't reliably hot unplug all added usb devices and usb
# passthrough breaks live migration we disable usb hotplugging for now
#qemu_deviceadd($vmid, PVE::QemuServer::USB::print_usbdevice_full($conf, $deviceid, $device));
-
} elsif ($deviceid =~ m/^(virtio)(\d+)$/) {
-
qemu_iothread_add($vmid, $deviceid, $device);
- qemu_driveadd($storecfg, $vmid, $device);
- my $devicefull = print_drivedevice_full($storecfg, $conf, $vmid, $device, undef, $arch, $machine_type);
+ qemu_driveadd($storecfg, $vmid, $device);
+ my $devicefull = print_drivedevice_full($storecfg, $conf, $vmid, $device, undef, $arch, $machine_type);
- qemu_deviceadd($vmid, $devicefull);
+ qemu_deviceadd($vmid, $devicefull);
eval { qemu_deviceaddverify($vmid, $deviceid); };
if (my $err = $@) {
eval { qemu_drivedel($vmid, $deviceid); };
warn $@ if $@;
die $err;
}
-
} elsif ($deviceid =~ m/^(virtioscsi|scsihw)(\d+)$/) {
-
-
- my $scsihw = defined($conf->{scsihw}) ? $conf->{scsihw} : "lsi";
- my $pciaddr = print_pci_addr($deviceid, undef, $arch, $machine_type);
+ my $scsihw = defined($conf->{scsihw}) ? $conf->{scsihw} : "lsi";
+ my $pciaddr = print_pci_addr($deviceid, undef, $arch, $machine_type);
my $scsihw_type = $scsihw eq 'virtio-scsi-single' ? "virtio-scsi-pci" : $scsihw;
- my $devicefull = "$scsihw_type,id=$deviceid$pciaddr";
+ my $devicefull = "$scsihw_type,id=$deviceid$pciaddr";
if($deviceid =~ m/^virtioscsi(\d+)$/ && $device->{iothread}) {
qemu_iothread_add($vmid, $deviceid, $device);
$devicefull .= ",num_queues=$device->{queues}";
}
- qemu_deviceadd($vmid, $devicefull);
- qemu_deviceaddverify($vmid, $deviceid);
-
+ qemu_deviceadd($vmid, $devicefull);
+ qemu_deviceaddverify($vmid, $deviceid);
} elsif ($deviceid =~ m/^(scsi)(\d+)$/) {
-
qemu_findorcreatescsihw($storecfg,$conf, $vmid, $device, $arch, $machine_type);
qemu_driveadd($storecfg, $vmid, $device);
warn $@ if $@;
die $err;
}
-
} elsif ($deviceid =~ m/^(net)(\d+)$/) {
-
return if !qemu_netdevadd($vmid, $conf, $arch, $device, $deviceid);
my $machine_type = PVE::QemuServer::Machine::qemu_machine_pxe($vmid, $conf);
warn $@ if $@;
die $err;
}
-
} elsif (!$q35 && $deviceid =~ m/^(pci\.)(\d+)$/) {
-
my $bridgeid = $2;
my $pciaddr = print_pci_addr($deviceid, undef, $arch, $machine_type);
my $devicefull = "pci-bridge,id=pci.$bridgeid,chassis_nr=$bridgeid$pciaddr";
qemu_deviceadd($vmid, $devicefull);
qemu_deviceaddverify($vmid, $deviceid);
-
} else {
die "can't hotplug device '$deviceid'\n";
}
my $device = parse_drive($deviceid, $conf->{$deviceid});
qemu_devicedel($vmid, $deviceid);
+ qemu_devicedelverify($vmid, $deviceid);
qemu_drivedel($vmid, $deviceid);
qemu_deletescsihw($conf, $vmid, $deviceid);
'tags' => 1,
};
+for my $opt (keys %$confdesc_cloudinit) {
+ $fast_plug_option->{$opt} = 1;
+};
+
# hotplug changes in [PENDING]
# $selection hash can be used to only apply specified options, for
# example: { cores => 1 } (only apply changed 'cores')
die "skip\n" if !$hotplug_features->{memory};
PVE::QemuServer::Memory::qemu_memory_hotplug($vmid, $conf, $defaults, $opt);
} elsif ($opt eq 'cpuunits') {
- $cgroup->change_cpu_shares(undef, 1024);
+ $cgroup->change_cpu_shares(undef);
} elsif ($opt eq 'cpulimit') {
- $cgroup->change_cpu_quota(-1, 100000);
+ $cgroup->change_cpu_quota(undef, undef); # reset, cgroup module can better decide values
} else {
die "skip\n";
}
}
}
- my ($apply_pending_cloudinit, $apply_pending_cloudinit_done);
- $apply_pending_cloudinit = sub {
- return if $apply_pending_cloudinit_done; # once is enough
- $apply_pending_cloudinit_done = 1; # once is enough
-
- my ($key, $value) = @_;
-
- my @cloudinit_opts = keys %$confdesc_cloudinit;
- foreach my $opt (keys %{$conf->{pending}}) {
- next if !grep { $_ eq $opt } @cloudinit_opts;
- $conf->{$opt} = delete $conf->{pending}->{$opt};
- }
-
- my $pending_delete_hash = PVE::QemuConfig->parse_pending_delete($conf->{pending}->{delete});
- foreach my $opt (sort keys %$pending_delete_hash) {
- next if !grep { $_ eq $opt } @cloudinit_opts;
- PVE::QemuConfig->remove_from_pending_delete($conf, $opt);
- delete $conf->{$opt};
- }
-
- my $new_conf = { %$conf };
- $new_conf->{$key} = $value;
- PVE::QemuServer::Cloudinit::generate_cloudinitconfig($new_conf, $vmid);
- };
-
foreach my $opt (keys %{$conf->{pending}}) {
next if $selection && !$selection->{$opt};
my $value = $conf->{pending}->{$opt};
# some changes can be done without hotplug
my $drive = parse_drive($opt, $value);
if (drive_is_cloudinit($drive)) {
- &$apply_pending_cloudinit($opt, $value);
+ PVE::QemuServer::Cloudinit::generate_cloudinitconfig($conf, $vmid);
}
vmconfig_update_disk($storecfg, $conf, $hotplug_features->{disk},
$vmid, $opt, $value, $arch, $machine_type);
die "skip\n" if !$hotplug_features->{memory};
$value = PVE::QemuServer::Memory::qemu_memory_hotplug($vmid, $conf, $defaults, $opt, $value);
} elsif ($opt eq 'cpuunits') {
- $cgroup->change_cpu_shares($conf->{pending}->{$opt}, 1024);
+ my $new_cpuunits = PVE::CGroup::clamp_cpu_shares($conf->{pending}->{$opt}); #clamp
+ $cgroup->change_cpu_shares($new_cpuunits);
} elsif ($opt eq 'cpulimit') {
my $cpulimit = $conf->{pending}->{$opt} == 0 ? -1 : int($conf->{pending}->{$opt} * 100000);
$cgroup->change_cpu_quota($cpulimit, 100000);
+ } elsif ($opt eq 'agent') {
+ vmconfig_update_agent($conf, $opt, $value);
} else {
die "skip\n"; # skip non-hot-pluggable options
}
delete $conf->{pending}->{$opt};
}
}
-
PVE::QemuConfig->write_config($vmid, $conf);
+
+ if($hotplug_features->{cloudinit}) {
+ my $pending = PVE::QemuServer::Cloudinit::get_pending_config($conf, $vmid);
+ my $regenerate = undef;
+ for my $item (@$pending) {
+ $regenerate = 1 if defined($item->{delete}) or defined($item->{pending});
+ }
+ PVE::QemuServer::vmconfig_update_cloudinit_drive($storecfg, $conf, $vmid) if $regenerate;
+ }
}
sub try_deallocate_drive {
sub vmconfig_apply_pending {
my ($vmid, $conf, $storecfg, $errors) = @_;
+ return if !scalar(keys %{$conf->{pending}});
+
my $add_apply_error = sub {
my ($opt, $msg) = @_;
my $err_msg = "unable to apply pending change $opt : $msg";
PVE::QemuConfig->cleanup_pending($conf);
+ my $generate_cloudnit = undef;
+
foreach my $opt (keys %{$conf->{pending}}) { # add/change
next if $opt eq 'delete'; # just to be sure
eval {
if (my $err = $@) {
$add_apply_error->($opt, $err);
} else {
+
+ if (is_valid_drivename($opt)) {
+ my $drive = parse_drive($opt, $conf->{pending}->{$opt});
+ $generate_cloudnit = 1 if drive_is_cloudinit($drive);
+ }
+
$conf->{$opt} = delete $conf->{pending}->{$opt};
}
}
# write all changes at once to avoid unnecessary i/o
PVE::QemuConfig->write_config($vmid, $conf);
+ PVE::QemuServer::Cloudinit::generate_cloudinitconfig($conf, $vmid) if $generate_cloudnit;
}
sub vmconfig_update_net {
}
}
+sub vmconfig_update_agent {
+ my ($conf, $opt, $value) = @_;
+
+ die "skip\n" if !$conf->{$opt};
+
+ my $hotplug_options = { fstrim_cloned_disks => 1 };
+
+ my $old_agent = parse_guest_agent($conf);
+ my $agent = parse_guest_agent({$opt => $value});
+
+ for my $option (keys %$agent) { # added/changed options
+ next if defined($hotplug_options->{$option});
+ die "skip\n" if safe_string_ne($agent->{$option}, $old_agent->{$option});
+ }
+
+ for my $option (keys %$old_agent) { # removed options
+ next if defined($hotplug_options->{$option});
+ die "skip\n" if safe_string_ne($old_agent->{$option}, $agent->{$option});
+ }
+
+ return; # either no actual change (e.g., format string reordered) or just hotpluggable changes
+}
+
sub vmconfig_update_disk {
my ($storecfg, $conf, $hotplug, $vmid, $opt, $value, $arch, $machine_type) = @_;
vm_deviceplug($storecfg, $conf, $vmid, $opt, $drive, $arch, $machine_type);
}
+sub vmconfig_update_cloudinit_drive {
+ my ($storecfg, $conf, $vmid) = @_;
+
+ my $cloudinit_ds = undef;
+ my $cloudinit_drive = undef;
+
+ PVE::QemuConfig->foreach_volume($conf, sub {
+ my ($ds, $drive) = @_;
+ if (PVE::QemuServer::drive_is_cloudinit($drive)) {
+ $cloudinit_ds = $ds;
+ $cloudinit_drive = $drive;
+ }
+ });
+
+ return if !$cloudinit_drive;
+
+ PVE::QemuServer::Cloudinit::generate_cloudinitconfig($conf, $vmid);
+ my $running = PVE::QemuServer::check_running($vmid);
+
+ if ($running) {
+ my $path = PVE::Storage::path($storecfg, $cloudinit_drive->{file});
+ if ($path) {
+ mon_cmd($vmid, "eject", force => JSON::true, id => "$cloudinit_ds");
+ mon_cmd($vmid, "blockdev-change-medium", id => "$cloudinit_ds", filename => "$path");
+ }
+ }
+}
+
# called in locked context by incoming migration
sub vm_migrate_get_nbd_disks {
my ($storecfg, $conf, $replicated_volumes) = @_;
my ($ds, $drive) = @_;
return if drive_is_cdrom($drive);
+ return if $ds eq 'tpmstate0';
my $volid = $drive->{file};
sub vm_migrate_alloc_nbd_disks {
my ($storecfg, $vmid, $source_volumes, $storagemap) = @_;
- my $format = undef;
-
my $nbd = {};
foreach my $opt (sort keys %$source_volumes) {
- my ($volid, $storeid, $volname, $drive, $use_existing) = @{$source_volumes->{$opt}};
+ my ($volid, $storeid, $volname, $drive, $use_existing, $format) = @{$source_volumes->{$opt}};
if ($use_existing) {
$nbd->{$opt}->{drivestr} = print_drive($drive);
next;
}
- # If a remote storage is specified and the format of the original
- # volume is not available there, fall back to the default format.
- # Otherwise use the same format as the original.
+ # storage mapping + volname = regular migration
+ # storage mapping + format = remote migration
+ # order of precedence, filtered by whether storage supports it:
+ # 1. explicit requested format
+ # 2. format of current volume
+ # 3. default format of storage
if (!$storagemap->{identity}) {
- $storeid = map_storage($storagemap, $storeid);
+ $storeid = PVE::JSONSchema::map_id($storagemap, $storeid);
my ($defFormat, $validFormats) = PVE::Storage::storage_default_format($storecfg, $storeid);
- my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
- my $fileFormat = qemu_img_format($scfg, $volname);
- $format = (grep {$fileFormat eq $_} @{$validFormats}) ? $fileFormat : $defFormat;
+ if (!$format || !grep { $format eq $_ } @$validFormats) {
+ if ($volname) {
+ my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
+ my $fileFormat = qemu_img_format($scfg, $volname);
+ $format = $fileFormat
+ if grep { $fileFormat eq $_ } @$validFormats;
+ }
+ $format //= $defFormat;
+ }
} else {
+ # can't happen for remote migration, so $volname is always defined
my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
$format = qemu_img_format($scfg, $volname);
}
# network => CIDR of migration network
# type => secure/insecure - tunnel over encrypted connection or plain-text
# nbd_proto_version => int, 0 for TCP, 1 for UNIX
-# replicated_volumes = which volids should be re-used with bitmaps for nbd migration
+# replicated_volumes => which volids should be re-used with bitmaps for nbd migration
+# offline_volumes => new volids of offline migrated disks like tpmstate and cloudinit, not yet
+# contained in config
sub vm_start_nolock {
my ($storecfg, $vmid, $conf, $params, $migrate_opts) = @_;
# this way we can reuse the old ISO with the correct config
PVE::QemuServer::Cloudinit::generate_cloudinitconfig($conf, $vmid) if !$migratedfrom;
+ # override offline migrated volumes, conf is out of date still
+ if (my $offline_volumes = $migrate_opts->{offline_volumes}) {
+ for my $key (sort keys $offline_volumes->%*) {
+ my $parsed = parse_drive($key, $conf->{$key});
+ $parsed->{file} = $offline_volumes->{$key};
+ $conf->{$key} = print_drive($parsed);
+ }
+ }
+
my $defaults = load_defaults();
# set environment variable useful inside network script
$pci_devices->{$i} = parse_hostpci($dev);
}
- my $pci_id_list = [ map { $_->{id} } map { $_->{pciid}->@* } values $pci_devices->%* ];
+ # do not reserve pciid for mediated devices, sysfs will error out for duplicate assignment
+ my $real_pci_devices = [ grep { !(defined($_->{mdev}) && scalar($_->{pciid}->@*) == 1) } values $pci_devices->%* ];
+
+ # map to a flat list of pci ids
+ my $pci_id_list = [ map { $_->{id} } map { $_->{pciid}->@* } $real_pci_devices->@* ];
+
# reserve all PCI IDs before actually doing anything with them
PVE::QemuServer::PCI::reserve_pci_usage($pci_id_list, $vmid, $start_timeout);
eval {
+ my $uuid;
for my $id (sort keys %$pci_devices) {
my $d = $pci_devices->{$id};
for my $dev ($d->{pciid}->@*) {
- PVE::QemuServer::PCI::prepare_pci_device($vmid, $dev->{id}, $id, $d->{mdev});
+ my $info = PVE::QemuServer::PCI::prepare_pci_device($vmid, $dev->{id}, $id, $d->{mdev});
+
+ # nvidia grid needs the uuid of the mdev as qemu parameter
+ if ($d->{mdev} && !defined($uuid) && $info->{vendor} eq '10de') {
+ $uuid = PVE::QemuServer::PCI::generate_mdev_uuid($vmid, $id);
+ }
}
}
+ push @$cmd, '-uuid', $uuid if defined($uuid);
};
if (my $err = $@) {
- eval { PVE::QemuServer::PCI::remove_pci_reservation($pci_id_list) };
+ eval { cleanup_pci_devices($vmid, $conf) };
warn $@ if $@;
die $err;
}
};
# Issues with the above 'stop' not being fully completed are extremely rare, a very low
# timeout should be more than enough here...
- PVE::Systemd::wait_for_unit_removed("$vmid.scope", 5);
+ PVE::Systemd::wait_for_unit_removed("$vmid.scope", 20);
- my $cpuunits = get_cpuunits($conf);
+ my $cpuunits = PVE::CGroup::clamp_cpu_shares($conf->{cpuunits});
my %run_params = (
timeout => $statefile ? undef : $start_timeout,
);
if (PVE::CGroup::cgroup_mode() == 2) {
- $cpuunits = 10000 if $cpuunits >= 10000; # else we get an error
$systemd_properties{CPUWeight} = $cpuunits;
} else {
$systemd_properties{CPUShares} = $cpuunits;
my $exitcode = run_command($cmd, %run_params);
if ($exitcode) {
- warn "stopping swtpm instance (pid $tpmpid) due to QEMU startup error\n";
- kill 'TERM', $tpmpid if $tpmpid;
+ if ($tpmpid) {
+ warn "stopping swtpm instance (pid $tpmpid) due to QEMU startup error\n";
+ kill 'TERM', $tpmpid;
+ }
die "QEMU exited with code $exitcode\n";
}
};
if (my $err = $@) {
# deactivate volumes if start fails
eval { PVE::Storage::deactivate_volumes($storecfg, $vollist); };
- eval { PVE::QemuServer::PCI::remove_pci_reservation($pci_id_list) };
+ warn $@ if $@;
+ eval { cleanup_pci_devices($vmid, $conf) };
+ warn $@ if $@;
die "start failed: $err";
}
my ($storecfg, $vmid, $snapname) = @_;
my $conf = PVE::QemuConfig->load_config($vmid);
- my $forcemachine;
- my $forcecpu;
+ my ($forcemachine, $forcecpu);
if ($snapname) {
my $snapshot = $conf->{snapshots}->{$snapname};
die "snapshot '$snapname' does not exist\n" if !defined($snapshot);
my $defaults = load_defaults();
- my $cmd = config_to_command($storecfg, $vmid, $conf, $defaults,
- $forcemachine, $forcecpu);
+ my $cmd = config_to_command($storecfg, $vmid, $conf, $defaults, $forcemachine, $forcecpu);
return PVE::Tools::cmd2string($cmd);
}
return $vollist;
}
+sub cleanup_pci_devices {
+ my ($vmid, $conf) = @_;
+
+ foreach my $key (keys %$conf) {
+ next if $key !~ m/^hostpci(\d+)$/;
+ my $hostpciindex = $1;
+ my $uuid = PVE::SysFSTools::generate_mdev_uuid($vmid, $hostpciindex);
+ my $d = parse_hostpci($conf->{$key});
+ if ($d->{mdev}) {
+ # NOTE: avoid PVE::SysFSTools::pci_cleanup_mdev_device as it requires PCI ID and we
+ # don't want to break ABI just for this two liner
+ my $dev_sysfs_dir = "/sys/bus/mdev/devices/$uuid";
+ PVE::SysFSTools::file_write("$dev_sysfs_dir/remove", "1") if -e $dev_sysfs_dir;
+ }
+ }
+ PVE::QemuServer::PCI::remove_pci_reservation($vmid);
+}
+
sub vm_stop_cleanup {
my ($storecfg, $vmid, $conf, $keepActive, $apply_pending_changes) = @_;
unlink '/dev/shm/pve-shm-' . ($ivshmem->{name} // $vmid);
}
- my $ids = [];
- foreach my $key (keys %$conf) {
- next if $key !~ m/^hostpci(\d+)$/;
- my $hostpciindex = $1;
- my $d = parse_hostpci($conf->{$key});
- my $uuid = PVE::SysFSTools::generate_mdev_uuid($vmid, $hostpciindex);
-
- foreach my $pci (@{$d->{pciid}}) {
- my $pciid = $pci->{id};
- push @$ids, $pci->{id};
- PVE::SysFSTools::pci_cleanup_mdev_device($pciid, $uuid);
- }
- }
- PVE::QemuServer::PCI::remove_pci_reservation($ids);
+ cleanup_pci_devices($vmid, $conf);
vmconfig_apply_pending($vmid, $conf, $storecfg) if $apply_pending_changes;
};
my $restore_cleanup_oldconf = sub {
my ($storecfg, $vmid, $oldconf, $virtdev_hash) = @_;
+ my $kept_disks = {};
+
PVE::QemuConfig->foreach_volume($oldconf, sub {
my ($ds, $drive) = @_;
if (my $err = $@) {
warn $err;
}
+ } else {
+ $kept_disks->{$volid} = 1;
}
});
- # delete vmstate files, after the restore we have no snapshots anymore
- foreach my $snapname (keys %{$oldconf->{snapshots}}) {
+ # after the restore we have no snapshots anymore
+ for my $snapname (keys $oldconf->{snapshots}->%*) {
my $snap = $oldconf->{snapshots}->{$snapname};
if ($snap->{vmstate}) {
eval { PVE::Storage::vdisk_free($storecfg, $snap->{vmstate}); };
warn $err;
}
}
+
+ for my $volid (keys $kept_disks->%*) {
+ eval { PVE::Storage::volume_snapshot_delete($storecfg, $volid, $snapname); };
+ warn $@ if $@;
+ }
}
};
my $parse_backup_hints = sub {
my ($rpcenv, $user, $storecfg, $fh, $devinfo, $options) = @_;
- my $virtdev_hash = {};
+ my $check_storage = sub { # assert if an image can be allocate
+ my ($storeid, $scfg) = @_;
+ die "Content type 'images' is not available on storage '$storeid'\n"
+ if !$scfg->{content}->{images};
+ $rpcenv->check($user, "/storage/$storeid", ['Datastore.AllocateSpace'])
+ if $user ne 'root@pam';
+ };
+ my $virtdev_hash = {};
while (defined(my $line = <$fh>)) {
if ($line =~ m/^\#qmdump\#map:(\S+):(\S+):(\S*):(\S*):$/) {
my ($virtdev, $devname, $storeid, $format) = ($1, $2, $3, $4);
$devinfo->{$devname}->{format} = $format;
$devinfo->{$devname}->{storeid} = $storeid;
- # check permission on storage
- my $pool = $options->{pool}; # todo: do we need that?
- if ($user ne 'root@pam') {
- $rpcenv->check($user, "/storage/$storeid", ['Datastore.AllocateSpace']);
- }
+ my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
+ $check_storage->($storeid, $scfg); # permission and content type check
$virtdev_hash->{$virtdev} = $devinfo->{$devname};
} elsif ($line =~ m/^((?:ide|sata|scsi)\d+):\s*(.*)\s*$/) {
my $virtdev = $1;
my $drive = parse_drive($virtdev, $2);
+
if (drive_is_cloudinit($drive)) {
my ($storeid, $volname) = PVE::Storage::parse_volume_id($drive->{file});
$storeid = $options->{storage} if defined ($options->{storage});
my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
my $format = qemu_img_format($scfg, $volname); # has 'raw' fallback
+ $check_storage->($storeid, $scfg); # permission and content type check
+
$virtdev_hash->{$virtdev} = {
format => $format,
storeid => $storeid,
}
my $restore_deactivate_volumes = sub {
- my ($storecfg, $devinfo) = @_;
+ my ($storecfg, $virtdev_hash) = @_;
my $vollist = [];
- foreach my $devname (keys %$devinfo) {
- my $volid = $devinfo->{$devname}->{volid};
- push @$vollist, $volid if $volid;
+ for my $dev (values $virtdev_hash->%*) {
+ push $vollist->@*, $dev->{volid} if $dev->{volid};
}
- PVE::Storage::deactivate_volumes($storecfg, $vollist);
+ eval { PVE::Storage::deactivate_volumes($storecfg, $vollist); };
+ print STDERR $@ if $@;
};
my $restore_destroy_volumes = sub {
- my ($storecfg, $devinfo) = @_;
+ my ($storecfg, $virtdev_hash) = @_;
- foreach my $devname (keys %$devinfo) {
- my $volid = $devinfo->{$devname}->{volid};
- next if !$volid;
+ for my $dev (values $virtdev_hash->%*) {
+ my $volid = $dev->{volid} or next;
eval {
- if ($volid =~ m|^/|) {
- unlink $volid || die 'unlink failed\n';
- } else {
- PVE::Storage::vdisk_free($storecfg, $volid);
- }
+ PVE::Storage::vdisk_free($storecfg, $volid);
print STDERR "temporary volume '$volid' sucessfuly removed\n";
};
print STDERR "unable to cleanup '$volid' - $@" if $@;
}
};
+my $restore_merge_config = sub {
+ my ($filename, $backup_conf_raw, $override_conf) = @_;
+
+ my $backup_conf = parse_vm_config($filename, $backup_conf_raw);
+ for my $key (keys $override_conf->%*) {
+ $backup_conf->{$key} = $override_conf->{$key};
+ }
+
+ return $backup_conf;
+};
+
sub scan_volids {
my ($cfg, $vmid) = @_;
my $keyfile = PVE::Storage::PBSPlugin::pbs_encryption_key_file_name($storecfg, $storeid);
my $repo = PVE::PBSClient::get_repository($scfg);
+ my $namespace = $scfg->{namespace};
# This is only used for `pbs-restore` and the QEMU PBS driver (live-restore)
my $password = PVE::Storage::PBSPlugin::pbs_get_password($scfg, $storeid);
my $new_conf_raw = '';
my $rpcenv = PVE::RPCEnvironment::get();
- my $devinfo = {};
+ my $devinfo = {}; # info about drives included in backup
+ my $virtdev_hash = {}; # info about allocated drives
eval {
# enable interrupts
my $index = PVE::Tools::file_get_contents($index_fn);
$index = decode_json($index);
- # print Dumper($index);
foreach my $info (@{$index->{files}}) {
if ($info->{filename} =~ m/^(drive-\S+).img.fidx$/) {
my $devname = $1;
my $fh = IO::File->new($cfgfn, "r") ||
die "unable to read qemu-server.conf - $!\n";
- my $virtdev_hash = $parse_backup_hints->($rpcenv, $user, $storecfg, $fh, $devinfo, $options);
+ $virtdev_hash = $parse_backup_hints->($rpcenv, $user, $storecfg, $fh, $devinfo, $options);
# fixme: rate limit?
# for live-restore we only want to preload the efidisk and TPM state
next if $options->{live} && $virtdev ne 'efidisk0' && $virtdev ne 'tpmstate0';
+ my @ns_arg;
+ if (defined(my $ns = $scfg->{namespace})) {
+ @ns_arg = ('--ns', $ns);
+ }
+
my $pbs_restore_cmd = [
'/usr/bin/pbs-restore',
'--repository', $repo,
+ @ns_arg,
$pbs_backup_name,
"$d->{devname}.img.fidx",
$path,
my $err = $@;
if ($err || !$options->{live}) {
- $restore_deactivate_volumes->($storecfg, $devinfo);
+ $restore_deactivate_volumes->($storecfg, $virtdev_hash);
}
rmtree $tmpdir;
if ($err) {
- $restore_destroy_volumes->($storecfg, $devinfo);
+ $restore_destroy_volumes->($storecfg, $virtdev_hash);
die $err;
}
$new_conf_raw .= "\nlock: create";
}
- PVE::Tools::file_set_contents($conffile, $new_conf_raw);
-
- PVE::Cluster::cfs_update(); # make sure we read new file
+ my $new_conf = $restore_merge_config->($conffile, $new_conf_raw, $options->{override_conf});
+ PVE::QemuConfig->write_config($vmid, $new_conf);
eval { rescan($vmid, 1); };
warn $@ if $@;
# these special drives are already restored before start
delete $devinfo->{'drive-efidisk0'};
delete $devinfo->{'drive-tpmstate0-backup'};
- pbs_live_restore($vmid, $conf, $storecfg, $devinfo, $repo, $keyfile, $pbs_backup_name);
+
+ my $pbs_opts = {
+ repo => $repo,
+ keyfile => $keyfile,
+ snapshot => $pbs_backup_name,
+ namespace => $namespace,
+ };
+ pbs_live_restore($vmid, $conf, $storecfg, $devinfo, $pbs_opts);
PVE::QemuConfig->remove_lock($vmid, "create");
}
}
sub pbs_live_restore {
- my ($vmid, $conf, $storecfg, $restored_disks, $repo, $keyfile, $snap) = @_;
+ my ($vmid, $conf, $storecfg, $restored_disks, $opts) = @_;
print "starting VM for live-restore\n";
- print "repository: '$repo', snapshot: '$snap'\n";
+ print "repository: '$opts->{repo}', snapshot: '$opts->{snapshot}'\n";
my $pbs_backing = {};
for my $ds (keys %$restored_disks) {
$ds =~ m/^drive-(.*)$/;
my $confname = $1;
$pbs_backing->{$confname} = {
- repository => $repo,
- snapshot => $snap,
+ repository => $opts->{repo},
+ snapshot => $opts->{snapshot},
archive => "$ds.img.fidx",
};
- $pbs_backing->{$confname}->{keyfile} = $keyfile if -e $keyfile;
+ $pbs_backing->{$confname}->{keyfile} = $opts->{keyfile} if -e $opts->{keyfile};
+ $pbs_backing->{$confname}->{namespace} = $opts->{namespace} if defined($opts->{namespace});
my $drive = parse_drive($confname, $conf->{$confname});
print "restoring '$ds' to '$drive->{file}'\n";
my $err = $@;
if ($err) {
- warn "An error occured during live-restore: $err\n";
+ warn "An error occurred during live-restore: $err\n";
_do_vm_stop($storecfg, $vmid, 1, 1, 10, 0, 1);
die "live-restore failed\n";
}
my $oldtimeout;
my $timeout = 5;
- my $devinfo = {};
+ my $devinfo = {}; # info about drives included in backup
+ my $virtdev_hash = {}; # info about allocated drives
my $rpcenv = PVE::RPCEnvironment::get();
PVE::Tools::file_copy($fwcfgfn, "${pve_firewall_dir}/$vmid.fw");
}
- my $virtdev_hash = $parse_backup_hints->($rpcenv, $user, $cfg, $fh, $devinfo, $opts);
+ $virtdev_hash = $parse_backup_hints->($rpcenv, $user, $cfg, $fh, $devinfo, $opts);
foreach my $info (values %{$virtdev_hash}) {
my $storeid = $info->{storeid};
alarm($oldtimeout) if $oldtimeout;
- $restore_deactivate_volumes->($cfg, $devinfo);
+ $restore_deactivate_volumes->($cfg, $virtdev_hash);
close($fifofh) if $fifofh;
unlink $mapfifo;
rmtree $tmpdir;
if ($err) {
- $restore_destroy_volumes->($cfg, $devinfo);
+ $restore_destroy_volumes->($cfg, $virtdev_hash);
die $err;
}
- PVE::Tools::file_set_contents($conffile, $new_conf_raw);
-
- PVE::Cluster::cfs_update(); # make sure we read new file
+ my $new_conf = $restore_merge_config->($conffile, $new_conf_raw, $opts->{override_conf});
+ PVE::QemuConfig->write_config($vmid, $new_conf);
eval { rescan($vmid, 1); };
warn $@ if $@;
sub restore_tar_archive {
my ($archive, $vmid, $user, $opts) = @_;
+ if (scalar(keys $opts->{override_conf}->%*) > 0) {
+ my $keystring = join(' ', keys $opts->{override_conf}->%*);
+ die "cannot pass along options ($keystring) when restoring from tar archive\n";
+ }
+
if ($archive ne '-') {
my $firstfile = tar_archive_read_firstfile($archive);
die "ERROR: file '$archive' does not look like a QemuServer vzdump backup\n"
$src_path = PVE::Storage::path($storecfg, $src_volid, $snapname);
$src_is_iscsi = ($src_path =~ m|^iscsi://|);
$cachemode = 'none' if $src_scfg->{type} eq 'zfspool';
- } elsif (-f $src_volid) {
+ } elsif (-f $src_volid || -b $src_volid) {
$src_path = $src_volid;
if ($src_path =~ m/\.($PVE::QemuServer::Drive::QEMU_FORMAT_RE)$/) {
$src_format = $1;
if ($agent_running) {
print "freeze filesystem\n";
eval { mon_cmd($vmid, "guest-fsfreeze-freeze"); };
+ warn $@ if $@;
} else {
print "suspend vm\n";
eval { PVE::QemuServer::vm_suspend($vmid, 1); };
+ warn $@ if $@;
}
# if we clone a disk for a new target vm, we don't switch the disk
if ($agent_running) {
print "unfreeze filesystem\n";
eval { mon_cmd($vmid, "guest-fsfreeze-thaw"); };
+ warn $@ if $@;
} else {
print "resume vm\n";
- eval { PVE::QemuServer::vm_resume($vmid, 1, 1); };
+ eval { PVE::QemuServer::vm_resume($vmid, 1, 1); };
+ warn $@ if $@;
}
last;
}
sub clone_disk {
- my ($storecfg, $vmid, $running, $drivename, $drive, $snapname,
- $newvmid, $storage, $format, $full, $newvollist, $jobs, $completion, $qga, $bwlimit, $conf) = @_;
+ my ($storecfg, $source, $dest, $full, $newvollist, $jobs, $completion, $qga, $bwlimit) = @_;
+
+ my ($vmid, $running) = $source->@{qw(vmid running)};
+ my ($src_drivename, $drive, $snapname) = $source->@{qw(drivename drive snapname)};
+
+ my ($newvmid, $dst_drivename, $efisize) = $dest->@{qw(vmid drivename efisize)};
+ my ($storage, $format) = $dest->@{qw(storage format)};
+
+ my $use_drive_mirror = $full && $running && $src_drivename && !$snapname;
+
+ if ($src_drivename && $dst_drivename && $src_drivename ne $dst_drivename) {
+ die "cloning from/to EFI disk requires EFI disk\n"
+ if $src_drivename eq 'efidisk0' || $dst_drivename eq 'efidisk0';
+ die "cloning from/to TPM state requires TPM state\n"
+ if $src_drivename eq 'tpmstate0' || $dst_drivename eq 'tpmstate0';
+
+ # This would lead to two device nodes in QEMU pointing to the same backing image!
+ die "cannot change drive name when cloning disk from/to the same VM\n"
+ if $use_drive_mirror && $vmid == $newvmid;
+ }
+
+ die "cannot move TPM state while VM is running\n"
+ if $use_drive_mirror && $src_drivename eq 'tpmstate0';
my $newvolid;
+ print "create " . ($full ? 'full' : 'linked') . " clone of drive ";
+ print "$src_drivename " if $src_drivename;
+ print "($drive->{file})\n";
+
if (!$full) {
- print "create linked clone of drive $drivename ($drive->{file})\n";
$newvolid = PVE::Storage::vdisk_clone($storecfg, $drive->{file}, $newvmid, $snapname);
push @$newvollist, $newvolid;
} else {
my $dst_format = resolve_dst_disk_format($storecfg, $storeid, $volname, $format);
- print "create full clone of drive $drivename ($drive->{file})\n";
my $name = undef;
my $size = undef;
if (drive_is_cloudinit($drive)) {
}
$snapname = undef;
$size = PVE::QemuServer::Cloudinit::CLOUDINIT_DISK_SIZE;
- } elsif ($drivename eq 'efidisk0') {
- $size = get_efivars_size($conf);
- } elsif ($drivename eq 'tpmstate0') {
+ } elsif ($dst_drivename eq 'efidisk0') {
+ $size = $efisize or die "internal error - need to specify EFI disk size\n";
+ } elsif ($dst_drivename eq 'tpmstate0') {
+ $dst_format = 'raw';
$size = PVE::QemuServer::Drive::TPMSTATE_DISK_SIZE;
} else {
($size) = PVE::Storage::volume_size_info($storecfg, $drive->{file}, 10);
}
my $sparseinit = PVE::Storage::volume_has_feature($storecfg, 'sparseinit', $newvolid);
- if (!$running || $snapname) {
+ if ($use_drive_mirror) {
+ qemu_drive_mirror($vmid, $src_drivename, $newvolid, $newvmid, $sparseinit, $jobs,
+ $completion, $qga, $bwlimit);
+ } else {
# TODO: handle bwlimits
- if ($drivename eq 'efidisk0') {
+ if ($dst_drivename eq 'efidisk0') {
# the relevant data on the efidisk may be smaller than the source
# e.g. on RBD/ZFS, so we use dd to copy only the amount
# that is given by the OVMF_VARS.fd
- my $src_path = PVE::Storage::path($storecfg, $drive->{file});
+ my $src_path = PVE::Storage::path($storecfg, $drive->{file}, $snapname);
my $dst_path = PVE::Storage::path($storecfg, $newvolid);
+ my $src_format = (PVE::Storage::parse_volname($storecfg, $drive->{file}))[6];
+
# better for Ceph if block size is not too small, see bug #3324
my $bs = 1024*1024;
- run_command(['qemu-img', 'dd', '-n', '-O', $dst_format, "bs=$bs", "osize=$size",
- "if=$src_path", "of=$dst_path"]);
+ my $cmd = ['qemu-img', 'dd', '-n', '-O', $dst_format];
+
+ if ($src_format eq 'qcow2' && $snapname) {
+ die "cannot clone qcow2 EFI disk snapshot - requires QEMU >= 6.2\n"
+ if !min_version(kvm_user_version(), 6, 2);
+ push $cmd->@*, '-l', $snapname;
+ }
+ push $cmd->@*, "bs=$bs", "osize=$size", "if=$src_path", "of=$dst_path";
+ run_command($cmd);
} else {
qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit);
}
- } else {
-
- die "cannot move TPM state while VM is running\n" if $drivename eq 'tpmstate0';
-
- my $kvmver = get_running_qemu_version ($vmid);
- if (!min_version($kvmver, 2, 7)) {
- die "drive-mirror with iothread requires qemu version 2.7 or higher\n"
- if $drive->{iothread};
- }
-
- qemu_drive_mirror($vmid, $drivename, $newvolid, $newvmid, $sparseinit, $jobs,
- $completion, $qga, $bwlimit);
}
}
no_data_clone:
my ($size) = eval { PVE::Storage::volume_size_info($storecfg, $newvolid, 10) };
- my $disk = $drive;
- $disk->{format} = undef;
+ my $disk = dclone($drive);
+ delete $disk->{format};
$disk->{file} = $newvolid;
$disk->{size} = $size if defined($size);
}
sub get_efivars_size {
- my ($conf) = @_;
+ my ($conf, $efidisk) = @_;
+
my $arch = get_vm_arch($conf);
- my $efidisk = $conf->{efidisk0} ? parse_drive('efidisk0', $conf->{efidisk0}) : undef;
- my (undef, $ovmf_vars) = get_ovmf_files($arch, $efidisk);
+ $efidisk //= $conf->{efidisk0} ? parse_drive('efidisk0', $conf->{efidisk0}) : undef;
+ my $smm = PVE::QemuServer::Machine::machine_type_is_q35($conf);
+ my (undef, $ovmf_vars) = get_ovmf_files($arch, $efidisk, $smm);
die "uefi vars image '$ovmf_vars' not found\n" if ! -f $ovmf_vars;
return -s $ovmf_vars;
}
$conf->{tpmstate0} = print_drive($disk);
}
-sub create_efidisk($$$$$$) {
- my ($storecfg, $storeid, $vmid, $fmt, $arch, $efidisk) = @_;
+sub create_efidisk($$$$$$$) {
+ my ($storecfg, $storeid, $vmid, $fmt, $arch, $efidisk, $smm) = @_;
- my (undef, $ovmf_vars) = get_ovmf_files($arch, $efidisk);
+ my (undef, $ovmf_vars) = get_ovmf_files($arch, $efidisk, $smm);
die "EFI vars default image not found\n" if ! -f $ovmf_vars;
my $vars_size_b = -s $ovmf_vars;
return ($maxdev, $controller, $controller_prefix);
}
-sub windows_version {
- my ($ostype) = @_;
-
- return 0 if !$ostype;
-
- my $winversion = 0;
-
- if($ostype eq 'wxp' || $ostype eq 'w2k3' || $ostype eq 'w2k') {
- $winversion = 5;
- } elsif($ostype eq 'w2k8' || $ostype eq 'wvista') {
- $winversion = 6;
- } elsif ($ostype =~ m/^win(\d+)$/) {
- $winversion = $1;
- }
-
- return $winversion;
-}
-
sub resolve_dst_disk_format {
my ($storecfg, $storeid, $src_volname, $format) = @_;
my ($defFormat, $validFormats) = PVE::Storage::storage_default_format($storecfg, $storeid);
projects / qemu-server.git / blobdiff