migrate: log which local resource causes error

[qemu-server.git] / PVE / QemuServer.pm

diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index 1ccdccf9410eca194366b0d2ca7be7ca83aa3078..9d560ec3a91031d95fe6fd056e3f2c8298514a7c 100644 (file)
--- a/PVE/QemuServer.pm
+++ b/PVE/QemuServer.pm
@@ -2,6 +2,7 @@ package PVE::QemuServer;
 
 use strict;
 use warnings;
 
 use strict;
 use warnings;

+

 use POSIX;
 use IO::Handle;
 use IO::Select;
 use POSIX;
 use IO::Handle;
 use IO::Select;


@@ -30,6 +31,7 @@ use PVE::ProcFSTools;
 use PVE::QemuConfig;
 use PVE::QMPClient;
 use PVE::RPCEnvironment;
 use PVE::QemuConfig;
 use PVE::QMPClient;
 use PVE::RPCEnvironment;

+use PVE::GuestHelpers;

 use PVE::QemuServer::PCI qw(print_pci_addr print_pcie_addr);
 use PVE::QemuServer::Memory;
 use PVE::QemuServer::USB qw(parse_usb_device);
 use PVE::QemuServer::PCI qw(print_pci_addr print_pcie_addr);
 use PVE::QemuServer::Memory;
 use PVE::QemuServer::USB qw(parse_usb_device);


@@ -76,12 +78,6 @@ PVE::JSONSchema::register_standard_option('pve-qm-stateuri', {
     optional => 1,
 });
 
     optional => 1,
 });
 

-PVE::JSONSchema::register_standard_option('pve-snapshot-name', {
-    description => "The name of the snapshot.",
-    type => 'string', format => 'pve-configid',
-    maxLength => 40,
-});
-

 PVE::JSONSchema::register_standard_option('pve-qm-image-format', {
     type => 'string',
     enum => [qw(raw cow qcow qed qcow2 vmdk cloop)],
 PVE::JSONSchema::register_standard_option('pve-qm-image-format', {
     type => 'string',
     enum => [qw(raw cow qcow qed qcow2 vmdk cloop)],


@@ -186,6 +182,13 @@ my $cpu_fmt = {
        optional => 1,
        default => 0
     },
        optional => 1,
        default => 0
     },

+    'hv-vendor-id' => {
+       type => 'string',
+       pattern => qr/[a-zA-Z0-9]{1,12}/,
+       format_description => 'vendor-id',
+       description => 'The Hyper-V vendor ID. Some drivers or programs inside Windows guests need a specific ID.',
+       optional => 1,
+    },

     flags => {
        description => "List of additional CPU flags separated by ';'."
                     . " Use '+FLAG' to enable, '-FLAG' to disable a flag."
     flags => {
        description => "List of additional CPU flags separated by ';'."
                     . " Use '+FLAG' to enable, '-FLAG' to disable a flag."


@@ -248,6 +251,21 @@ my $vga_fmt = {
     },
 };
 
     },
 };
 

+my $ivshmem_fmt = {
+    size => {
+       type => 'integer',
+       minimum => 1,
+       description => "The size of the file in MB.",
+    },
+    name => {
+       type => 'string',
+       pattern => '[a-zA-Z0-9\-]+',
+       optional => 1,
+       format_description => 'string',
+       description => "The name of the file. Will be prefixed with 'pve-shm-'. Default is the VMID. Will be deleted when the VM is stopped.",
+    },
+};
+

 my $confdesc = {
     onboot => {
        optional => 1,
 my $confdesc = {
     onboot => {
        optional => 1,


@@ -277,7 +295,7 @@ my $confdesc = {
        optional => 1,
        type => 'string',
        description => "Lock/unlock the VM.",
        optional => 1,
        type => 'string',
        description => "Lock/unlock the VM.",

-       enum => [qw(backup clone create migrate rollback snapshot snapshot-delete)],
+       enum => [qw(backup clone create migrate rollback snapshot snapshot-delete suspending suspended)],

     },
     cpulimit => {
        optional => 1,
     },
     cpulimit => {
        optional => 1,


@@ -607,8 +625,45 @@ EODESCR
        default => "1 (autogenerated)",
        optional => 1,
     },
        default => "1 (autogenerated)",
        optional => 1,
     },

+    hookscript => {
+       type => 'string',
+       format => 'pve-volume-id',
+       optional => 1,
+       description => "Script that will be executed during various steps in the vms lifetime.",
+    },
+    ivshmem => {
+       type => 'string',
+       format => $ivshmem_fmt,
+       description => "Inter-VM shared memory. Useful for direct communication between VMs, or to the host.",
+       optional => 1,
+    }

 };
 
 };
 

+my $cicustom_fmt = {
+    meta => {
+       type => 'string',
+       optional => 1,
+       description => 'Specify a custom file containing all meta data passed to the VM via cloud-init. This is provider specific meaning configdrive2 and nocloud differ.',
+       format => 'pve-volume-id',
+       format_description => 'volume',
+    },
+    network => {
+       type => 'string',
+       optional => 1,
+       description => 'Specify a custom file containing all network data passed to the VM via cloud-init.',
+       format => 'pve-volume-id',
+       format_description => 'volume',
+    },
+    user => {
+       type => 'string',
+       optional => 1,
+       description => 'Specify a custom file containing all user data passed to the VM via cloud-init.',
+       format => 'pve-volume-id',
+       format_description => 'volume',
+    },
+};
+PVE::JSONSchema::register_format('pve-qm-cicustom', $cicustom_fmt);
+

 my $confdesc_cloudinit = {
     citype => {
        optional => 1,
 my $confdesc_cloudinit = {
     citype => {
        optional => 1,


@@ -626,6 +681,12 @@ my $confdesc_cloudinit = {
        type => 'string',
        description => 'cloud-init: Password to assign the user. Using this is generally not recommended. Use ssh keys instead. Also note that older cloud-init versions do not support hashed passwords.',
     },
        type => 'string',
        description => 'cloud-init: Password to assign the user. Using this is generally not recommended. Use ssh keys instead. Also note that older cloud-init versions do not support hashed passwords.',
     },

+    cicustom => {
+       optional => 1,
+       type => 'string',
+       description => 'cloud-init: Specify custom files to replace the automatically generated ones at start.',
+       format => 'pve-qm-cicustom',
+    },

     searchdomain => {
        optional => 1,
        type => 'string',
     searchdomain => {
        optional => 1,
        type => 'string',


@@ -736,13 +797,9 @@ The DHCP server assign addresses to the guest starting from 10.0.2.15.
 __EOD__
 
 my $net_fmt = {
 __EOD__
 
 my $net_fmt = {

-    macaddr => {
-       type => 'string',
-       pattern => qr/[0-9a-f]{2}(?::[0-9a-f]{2}){5}/i,
+    macaddr  => get_standard_option('mac-addr', {

        description => "MAC address. That address must be unique withing your network. This is automatically generated if not specified.",
        description => "MAC address. That address must be unique withing your network. This is automatically generated if not specified.",

-       format_description => "XX:XX:XX:XX:XX:XX",
-       optional => 1,
-    },
+    }),

     model => {
        type => 'string',
        description => "Network Card Model. The 'virtio' model provides the best performance with very low CPU overhead. If your guest does not support this driver, it is usually best to use 'e1000'.",
     model => {
        type => 'string',
        description => "Network Card Model. The 'virtio' model provides the best performance with very low CPU overhead. If your guest does not support this driver, it is usually best to use 'e1000'.",


@@ -1038,6 +1095,16 @@ my %ssd_fmt = (
     },
 );
 
     },
 );
 

+my %wwn_fmt = (
+    wwn => {
+       type => 'string',
+       pattern => qr/^(0x)[0-9a-fA-F]{16}/,
+       format_description => 'wwn',
+       description => "The drive's worldwide name, encoded as 16 bytes hex string, prefixed by '0x'.",
+       optional => 1,
+    },
+);
+

 my $add_throttle_desc = sub {
     my ($key, $type, $what, $unit, $longunit, $minimum) = @_;
     my $d = {
 my $add_throttle_desc = sub {
     my ($key, $type, $what, $unit, $longunit, $minimum) = @_;
     my $d = {


@@ -1086,6 +1153,7 @@ my $ide_fmt = {
     %drivedesc_base,
     %model_fmt,
     %ssd_fmt,
     %drivedesc_base,
     %model_fmt,
     %ssd_fmt,

+    %wwn_fmt,

 };
 PVE::JSONSchema::register_format("pve-qm-ide", $ide_fmt);
 
 };
 PVE::JSONSchema::register_format("pve-qm-ide", $ide_fmt);
 


@@ -1102,6 +1170,7 @@ my $scsi_fmt = {
     %queues_fmt,
     %scsiblock_fmt,
     %ssd_fmt,
     %queues_fmt,
     %scsiblock_fmt,
     %ssd_fmt,

+    %wwn_fmt,

 };
 my $scsidesc = {
     optional => 1,
 };
 my $scsidesc = {
     optional => 1,


@@ -1113,6 +1182,7 @@ PVE::JSONSchema::register_standard_option("pve-qm-scsi", $scsidesc);
 my $sata_fmt = {
     %drivedesc_base,
     %ssd_fmt,
 my $sata_fmt = {
     %drivedesc_base,
     %ssd_fmt,

+    %wwn_fmt,

 };
 my $satadesc = {
     optional => 1,
 };
 my $satadesc = {
     optional => 1,


@@ -1139,6 +1209,7 @@ my $alldrive_fmt = {
     %queues_fmt,
     %scsiblock_fmt,
     %ssd_fmt,
     %queues_fmt,
     %scsiblock_fmt,
     %ssd_fmt,

+    %wwn_fmt,

 };
 
 my $efidisk_fmt = {
 };
 
 my $efidisk_fmt = {


@@ -1210,7 +1281,7 @@ my $hostpci_fmt = {
        pattern => qr/$PCIRE(;$PCIRE)*/,
        format_description => 'HOSTPCIID[;HOSTPCIID2...]',
        description => <<EODESCR,
        pattern => qr/$PCIRE(;$PCIRE)*/,
        format_description => 'HOSTPCIID[;HOSTPCIID2...]',
        description => <<EODESCR,

-Host PCI device pass through. The PCI ID of a host's PCI device or a list 
+Host PCI device pass through. The PCI ID of a host's PCI device or a list

 of PCI virtual functions of the host. HOSTPCIID syntax is:
 
 'bus:dev.func' (hexadecimal numbers)
 of PCI virtual functions of the host. HOSTPCIID syntax is:
 
 'bus:dev.func' (hexadecimal numbers)


@@ -1264,7 +1335,7 @@ my $hostpcidesc = {
        verbose_description =>  <<EODESCR,
 Map host PCI devices into guest.
 
        verbose_description =>  <<EODESCR,
 Map host PCI devices into guest.
 

-NOTE: This option allows direct access to host hardware. So it is no longer 
+NOTE: This option allows direct access to host hardware. So it is no longer

 possible to migrate such machines - use with special care.
 
 CAUTION: Experimental! User reported problems with this option.
 possible to migrate such machines - use with special care.
 
 CAUTION: Experimental! User reported problems with this option.


@@ -1770,6 +1841,7 @@ sub print_drivedevice_full {
        if ($drive->{ssd} && ($devicetype eq 'block' || $devicetype eq 'hd')) {
            $device .= ",rotation_rate=1";
        }
        if ($drive->{ssd} && ($devicetype eq 'block' || $devicetype eq 'hd')) {
            $device .= ",rotation_rate=1";
        }

+       $device .= ",wwn=$drive->{wwn}" if $drive->{wwn};

 
     } elsif ($drive->{interface} eq 'ide' || $drive->{interface} eq 'sata') {
        my $maxdev = ($drive->{interface} eq 'sata') ? $MAX_SATA_DISKS : 2;
 
     } elsif ($drive->{interface} eq 'ide' || $drive->{interface} eq 'sata') {
        my $maxdev = ($drive->{interface} eq 'sata') ? $MAX_SATA_DISKS : 2;


@@ -1794,6 +1866,7 @@ sub print_drivedevice_full {
                $device .= ",rotation_rate=1";
            }
        }
                $device .= ",rotation_rate=1";
            }
        }

+       $device .= ",wwn=$drive->{wwn}" if $drive->{wwn};

     } elsif ($drive->{interface} eq 'usb') {
        die "implement me";
        #  -device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0
     } elsif ($drive->{interface} eq 'usb') {
        die "implement me";
        #  -device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0


@@ -1832,7 +1905,7 @@ sub print_drive_full {
     my $path;
     my $volid = $drive->{file};
     my $format;
     my $path;
     my $volid = $drive->{file};
     my $format;

-    
+

     if (drive_is_cdrom($drive)) {
        $path = get_iso_path($storecfg, $vmid, $volid);
     } else {
     if (drive_is_cdrom($drive)) {
        $path = get_iso_path($storecfg, $vmid, $volid);
     } else {


@@ -2773,21 +2846,23 @@ sub config_list {
 sub check_local_resources {
     my ($conf, $noerr) = @_;
 
 sub check_local_resources {
     my ($conf, $noerr) = @_;
 

-    my $loc_res = 0;
+    my @loc_res = ();

 
 

-    $loc_res = 1 if $conf->{hostusb}; # old syntax
-    $loc_res = 1 if $conf->{hostpci}; # old syntax
+    push @loc_res, "hostusb" if $conf->{hostusb}; # old syntax
+    push @loc_res, "hostpci" if $conf->{hostpci}; # old syntax
+
+    push @loc_res, "ivshmem" if $conf->{ivshmem};

 
     foreach my $k (keys %$conf) {
        next if $k =~ m/^usb/ && ($conf->{$k} eq 'spice');
        # sockets are safe: they will recreated be on the target side post-migrate
        next if $k =~ m/^serial/ && ($conf->{$k} eq 'socket');
 
     foreach my $k (keys %$conf) {
        next if $k =~ m/^usb/ && ($conf->{$k} eq 'spice');
        # sockets are safe: they will recreated be on the target side post-migrate
        next if $k =~ m/^serial/ && ($conf->{$k} eq 'socket');

-       $loc_res = 1 if $k =~ m/^(usb|hostpci|serial|parallel)\d+$/;
+       push @loc_res, $k if $k =~ m/^(usb|hostpci|serial|parallel)\d+$/;

     }
 
     }
 

-    die "VM uses local resources\n" if $loc_res && !$noerr;
+    die "VM uses local resources\n" if scalar @loc_res && !$noerr;

 
 

-    return $loc_res;
+    return \@loc_res;

 }
 
 # check if used storages are available on all nodes (use by migrate)
 }
 
 # check if used storages are available on all nodes (use by migrate)


@@ -2985,6 +3060,11 @@ our $vmstatus_return_properties = {
        type => 'number',
        optional => 1,
     },
        type => 'number',
        optional => 1,
     },

+    lock => {
+       description => "The current config lock, if any.",
+       type => 'string',
+       optional => 1,
+    }

 };
 
 my $last_proc_pid_stat;
 };
 
 my $last_proc_pid_stat;


@@ -3055,6 +3135,7 @@ sub vmstatus {
         $d->{template} = PVE::QemuConfig->is_template($conf);
 
        $d->{serial} = 1 if conf_has_serial($conf);
         $d->{template} = PVE::QemuConfig->is_template($conf);
 
        $d->{serial} = 1 if conf_has_serial($conf);

+       $d->{lock} = $conf->{lock} if $conf->{lock};

 
        $res->{$vmid} = $d;
     }
 
        $res->{$vmid} = $d;
     }


@@ -3331,11 +3412,13 @@ sub get_cpu_options {
     if ($arch eq 'aarch64') {
        $cpu = 'cortex-a57';
     }
     if ($arch eq 'aarch64') {
        $cpu = 'cortex-a57';
     }

+    my $hv_vendor_id;

     if (my $cputype = $conf->{cpu}) {
        my $cpuconf = PVE::JSONSchema::parse_property_string($cpu_fmt, $cputype)
            or die "Cannot parse cpu description: $cputype\n";
        $cpu = $cpuconf->{cputype};
        $kvm_off = 1 if $cpuconf->{hidden};
     if (my $cputype = $conf->{cpu}) {
        my $cpuconf = PVE::JSONSchema::parse_property_string($cpu_fmt, $cputype)
            or die "Cannot parse cpu description: $cputype\n";
        $cpu = $cpuconf->{cputype};
        $kvm_off = 1 if $cpuconf->{hidden};

+       $hv_vendor_id = $cpuconf->{'hv-vendor-id'};

 
        if (defined(my $flags = $cpuconf->{flags})) {
            push @$cpuFlags, split(";", $flags);
 
        if (defined(my $flags = $cpuconf->{flags})) {
            push @$cpuFlags, split(";", $flags);


@@ -3357,7 +3440,7 @@ sub get_cpu_options {
        push @$cpuFlags , '+kvm_pv_eoi' if $kvm;
     }
 
        push @$cpuFlags , '+kvm_pv_eoi' if $kvm;
     }
 

-    add_hyperv_enlightenments($cpuFlags, $winversion, $machine_type, $kvmver, $conf->{bios}, $gpu_passthrough) if $kvm;
+    add_hyperv_enlightenments($cpuFlags, $winversion, $machine_type, $kvmver, $conf->{bios}, $gpu_passthrough, $hv_vendor_id) if $kvm;

 
     push @$cpuFlags, 'enforce' if $cpu ne 'host' && $kvm && $arch eq 'x86_64';
 
 
     push @$cpuFlags, 'enforce' if $cpu ne 'host' && $kvm && $arch eq 'x86_64';
 


@@ -3524,7 +3607,7 @@ sub config_to_command {
        next if !$d;
 
        my $pcie = $d->{pcie};
        next if !$d;
 
        my $pcie = $d->{pcie};

-       if($pcie){
+       if ($pcie) {

            die "q35 machine model is not enabled" if !$q35;
            # win7 wants to have the pcie devices directly on the pcie bus
            # instead of in the root port
            die "q35 machine model is not enabled" if !$q35;
            # win7 wants to have the pcie devices directly on the pcie bus
            # instead of in the root port


@@ -3533,7 +3616,7 @@ sub config_to_command {
            } else {
                $pciaddr = print_pcie_addr("hostpci$i");
            }
            } else {
                $pciaddr = print_pcie_addr("hostpci$i");
            }

-       }else{
+       } else {

            $pciaddr = print_pci_addr("hostpci$i", $bridges, $arch, $machine_type);
        }
 
            $pciaddr = print_pci_addr("hostpci$i", $bridges, $arch, $machine_type);
        }
 


@@ -3544,7 +3627,7 @@ sub config_to_command {
        if ($d->{'x-vga'}) {
            $xvga = ',x-vga=on';
            $kvm_off = 1;
        if ($d->{'x-vga'}) {
            $xvga = ',x-vga=on';
            $kvm_off = 1;

-           $vga->{type} = 'none';
+           $vga->{type} = 'none' if !defined($conf->{vga});

            $gpu_passthrough = 1;
 
            if ($conf->{bios} && $conf->{bios} eq 'ovmf') {
            $gpu_passthrough = 1;
 
            if ($conf->{bios} && $conf->{bios} eq 'ovmf') {


@@ -3715,7 +3798,7 @@ sub config_to_command {
     push @$cmd, get_cpu_options($conf, $arch, $kvm, $machine_type, $kvm_off, $kvmver, $winversion, $gpu_passthrough);
 
     PVE::QemuServer::Memory::config($conf, $vmid, $sockets, $cores, $defaults, $hotplug_features, $cmd);
     push @$cmd, get_cpu_options($conf, $arch, $kvm, $machine_type, $kvm_off, $kvmver, $winversion, $gpu_passthrough);
 
     PVE::QemuServer::Memory::config($conf, $vmid, $sockets, $cores, $defaults, $hotplug_features, $cmd);

-    
+

     push @$cmd, '-S' if $conf->{freeze};
 
     push @$cmd, '-k', $conf->{keyboard} if defined($conf->{keyboard});
     push @$cmd, '-S' if $conf->{freeze};
 
     push @$cmd, '-k', $conf->{keyboard} if defined($conf->{keyboard});


@@ -3839,7 +3922,7 @@ sub config_to_command {
            my $queues = '';
            if($conf->{scsihw} && $conf->{scsihw} eq "virtio-scsi-single" && $drive->{queues}){
                $queues = ",num_queues=$drive->{queues}";
            my $queues = '';
            if($conf->{scsihw} && $conf->{scsihw} eq "virtio-scsi-single" && $drive->{queues}){
                $queues = ",num_queues=$drive->{queues}";

-           } 
+           }

 
            push @$devices, '-device', "$scsihw_type,id=$controller_prefix$controller$pciaddr$iothread$queues" if !$scsicontroller->{$controller};
            $scsicontroller->{$controller}=1;
 
            push @$devices, '-device', "$scsihw_type,id=$controller_prefix$controller$pciaddr$iothread$queues" if !$scsicontroller->{$controller};
            $scsicontroller->{$controller}=1;


@@ -3876,6 +3959,23 @@ sub config_to_command {
          push @$devices, '-device', $netdevicefull;
     }
 
          push @$devices, '-device', $netdevicefull;
     }
 

+    if ($conf->{ivshmem}) {
+       my $ivshmem = PVE::JSONSchema::parse_property_string($ivshmem_fmt, $conf->{ivshmem});
+
+       my $bus;
+       if ($q35) {
+           $bus = print_pcie_addr("ivshmem");
+       } else {
+           $bus = print_pci_addr("ivshmem", $bridges, $arch, $machine_type);
+       }
+
+       my $ivshmem_name = $ivshmem->{name} // $vmid;
+       my $path = '/dev/shm/pve-shm-' . $ivshmem_name;
+
+       push @$devices, '-device', "ivshmem-plain,memdev=ivshmem$bus,";
+       push @$devices, '-object', "memory-backend-file,id=ivshmem,share=on,mem-path=$path,size=$ivshmem->{size}M";
+    }
+

     if (!$q35) {
        # add pci bridges
         if (qemu_machine_feature_enabled ($machine_type, $kvmver, 2, 3)) {
     if (!$q35) {
        # add pci bridges
         if (qemu_machine_feature_enabled ($machine_type, $kvmver, 2, 3)) {


@@ -3899,6 +3999,12 @@ sub config_to_command {
     push @$cmd, '-global', join(',', @$globalFlags)
        if scalar(@$globalFlags);
 
     push @$cmd, '-global', join(',', @$globalFlags)
        if scalar(@$globalFlags);
 

+    if (my $vmstate = $conf->{vmstate}) {
+       my $statepath = PVE::Storage::path($storecfg, $vmstate);
+       PVE::Storage::activate_volumes($storecfg, [$vmstate]);
+       push @$cmd, '-loadstate', $statepath;
+    }
+

     # add custom args
     if ($conf->{args}) {
        my $aa = PVE::Tools::split_args($conf->{args});
     # add custom args
     if ($conf->{args}) {
        my $aa = PVE::Tools::split_args($conf->{args});


@@ -4060,20 +4166,23 @@ sub vm_deviceplug {
 
     } elsif ($deviceid =~ m/^(net)(\d+)$/) {
 
 
     } elsif ($deviceid =~ m/^(net)(\d+)$/) {
 

-        return undef if !qemu_netdevadd($vmid, $conf, $arch, $device, $deviceid);
+       return undef if !qemu_netdevadd($vmid, $conf, $arch, $device, $deviceid);

 
 

-        my $machine_type = PVE::QemuServer::qemu_machine_pxe($vmid, $conf); 
-        my $use_old_bios_files = undef;
-        ($use_old_bios_files, $machine_type) = qemu_use_old_bios_files($machine_type);
+       my $machine_type = PVE::QemuServer::qemu_machine_pxe($vmid, $conf);
+       my $use_old_bios_files = undef;
+       ($use_old_bios_files, $machine_type) = qemu_use_old_bios_files($machine_type);

 
 

-        my $netdevicefull = print_netdevice_full($vmid, $conf, $device, $deviceid, undef, $use_old_bios_files, $arch, $machine_type);
-        qemu_deviceadd($vmid, $netdevicefull);
-        eval { qemu_deviceaddverify($vmid, $deviceid); };
+       my $netdevicefull = print_netdevice_full($vmid, $conf, $device, $deviceid, undef, $use_old_bios_files, $arch, $machine_type);
+       qemu_deviceadd($vmid, $netdevicefull);
+       eval {
+           qemu_deviceaddverify($vmid, $deviceid);
+           qemu_set_link_status($vmid, $deviceid, !$device->{link_down});
+       };

        if (my $err = $@) {
            eval { qemu_netdevdel($vmid, $deviceid); };
            warn $@ if $@;
            die $err;
        if (my $err = $@) {
            eval { qemu_netdevdel($vmid, $deviceid); };
            warn $@ if $@;
            die $err;

-        }
+       }

 
     } elsif (!$q35 && $deviceid =~ m/^(pci\.)(\d+)$/) {
 
 
     } elsif (!$q35 && $deviceid =~ m/^(pci\.)(\d+)$/) {
 


@@ -4172,7 +4281,11 @@ sub qemu_iothread_add {
 sub qemu_iothread_del {
     my($conf, $vmid, $deviceid) = @_;
 
 sub qemu_iothread_del {
     my($conf, $vmid, $deviceid) = @_;
 

-    my $device = parse_drive($deviceid, $conf->{$deviceid});
+    my $confid = $deviceid;
+    if ($deviceid =~ m/^(?:virtioscsi|scsihw)(\d+)$/) {
+       $confid = 'scsi' . $1;
+    }
+    my $device = parse_drive($confid, $conf->{$confid});

     if ($device->{iothread}) {
        my $iothreads = vm_iothreads_list($vmid);
        qemu_objectdel($vmid, "iothread-$deviceid") if $iothreads->{"iothread-$deviceid"};
     if ($device->{iothread}) {
        my $iothreads = vm_iothreads_list($vmid);
        qemu_objectdel($vmid, "iothread-$deviceid") if $iothreads->{"iothread-$deviceid"};


@@ -4587,6 +4700,7 @@ my $fast_plug_option = {
     'description' => 1,
     'protection' => 1,
     'vmstatestorage' => 1,
     'description' => 1,
     'protection' => 1,
     'vmstatestorage' => 1,

+    'hookscript' => 1,

 };
 
 # hotplug changes in [PENDING]
 };
 
 # hotplug changes in [PENDING]


@@ -5041,7 +5155,10 @@ sub vm_start {
 
        die "you can't start a vm if it's a template\n" if PVE::QemuConfig->is_template($conf);
 
 
        die "you can't start a vm if it's a template\n" if PVE::QemuConfig->is_template($conf);
 

-       PVE::QemuConfig->check_lock($conf) if !$skiplock;
+       my $is_suspended = PVE::QemuConfig->has_lock($conf, 'suspended');
+
+       PVE::QemuConfig->check_lock($conf)
+           if !($skiplock || $is_suspended);

 
        die "VM $vmid already running\n" if check_running($vmid, undef, $migratedfrom);
 
 
        die "VM $vmid already running\n" if check_running($vmid, undef, $migratedfrom);
 


@@ -5105,6 +5222,14 @@ sub vm_start {
            }
        }
 
            }
        }
 

+       PVE::GuestHelpers::exec_hookscript($conf, $vmid, 'pre-start', 1);
+
+       if ($is_suspended) {
+           # enforce machine type on suspended vm to ensure HW compatibility
+           $forcemachine = $conf->{runningmachine};
+           print "Resuming suspended VM\n";
+       }
+

        my ($cmd, $vollist, $spice_port) = config_to_command($storecfg, $vmid, $conf, $defaults, $forcemachine);
 
        my $migrate_port = 0;
        my ($cmd, $vollist, $spice_port) = config_to_command($storecfg, $vmid, $conf, $defaults, $forcemachine);
 
        my $migrate_port = 0;


@@ -5185,7 +5310,7 @@ sub vm_start {
 
        PVE::Storage::activate_volumes($storecfg, $vollist);
 
 
        PVE::Storage::activate_volumes($storecfg, $vollist);
 

-       if (!check_running($vmid, 1)) {
+       if (-d "/sys/fs/cgroup/systemd/qemu.slice/$vmid.scope") {

            eval {
                run_command(['/bin/systemctl', 'stop', "$vmid.scope"],
                    outfunc => sub {}, errfunc => sub {});
            eval {
                run_command(['/bin/systemctl', 'stop', "$vmid.scope"],
                    outfunc => sub {}, errfunc => sub {});


@@ -5195,7 +5320,7 @@ sub vm_start {
        my $cpuunits = defined($conf->{cpuunits}) ? $conf->{cpuunits}
                                                  : $defaults->{cpuunits};
 
        my $cpuunits = defined($conf->{cpuunits}) ? $conf->{cpuunits}
                                                  : $defaults->{cpuunits};
 

-       my $start_timeout = $conf->{hugepages} ? 300 : 30;
+       my $start_timeout = ($conf->{hugepages} || $is_suspended) ? 300 : 30;

        my %run_params = (timeout => $statefile ? undef : $start_timeout, umask => 0077);
 
        my %properties = (
        my %run_params = (timeout => $statefile ? undef : $start_timeout, umask => 0077);
 
        my %properties = (


@@ -5302,6 +5427,15 @@ sub vm_start {
                    property => "guest-stats-polling-interval",
                    value => 2) if (!defined($conf->{balloon}) || $conf->{balloon});
 
                    property => "guest-stats-polling-interval",
                    value => 2) if (!defined($conf->{balloon}) || $conf->{balloon});
 

+       if ($is_suspended && (my $vmstate = $conf->{vmstate})) {
+           print "Resumed VM, removing state\n";
+           delete $conf->@{qw(lock vmstate runningmachine)};
+           PVE::Storage::deactivate_volumes($storecfg, [$vmstate]);
+           PVE::Storage::vdisk_free($storecfg, $vmstate);
+           PVE::QemuConfig->write_config($vmid, $conf);
+       }
+
+       PVE::GuestHelpers::exec_hookscript($conf, $vmid, 'post-start');

     });
 }
 
     });
 }
 


@@ -5363,10 +5497,19 @@ sub vm_human_monitor_command {
 }
 
 sub vm_commandline {
 }
 
 sub vm_commandline {

-    my ($storecfg, $vmid) = @_;
+    my ($storecfg, $vmid, $snapname) = @_;

 
     my $conf = PVE::QemuConfig->load_config($vmid);
 
 
     my $conf = PVE::QemuConfig->load_config($vmid);
 

+    if ($snapname) {
+       my $snapshot = $conf->{snapshots}->{$snapname};
+       die "snapshot '$snapname' does not exist\n" if !defined($snapshot);
+
+       $snapshot->{digest} = $conf->{digest}; # keep file digest for API
+
+       $conf = $snapshot;
+    }
+

     my $defaults = load_defaults();
 
     my $cmd = config_to_command($storecfg, $vmid, $conf, $defaults);
     my $defaults = load_defaults();
 
     my $cmd = config_to_command($storecfg, $vmid, $conf, $defaults);


@@ -5419,6 +5562,15 @@ sub vm_stop_cleanup {
            unlink "/var/run/qemu-server/${vmid}.$ext";
        }
 
            unlink "/var/run/qemu-server/${vmid}.$ext";
        }
 

+       if ($conf->{ivshmem}) {
+           my $ivshmem = PVE::JSONSchema::parse_property_string($ivshmem_fmt, $conf->{ivshmem});
+           # just delete it for now, VMs which have this already open do not
+           # are affected, but new VMs will get a separated one. If this
+           # becomes an issue we either add some sort of ref-counting or just
+           # add a "don't delete on stop" flag to the ivshmem format.
+           unlink '/dev/shm/pve-shm-' . ($ivshmem->{name} // $vmid);
+       }
+

        foreach my $key (keys %$conf) {
            next if $key !~ m/^hostpci(\d+)$/;
            my $hostpciindex = $1;
        foreach my $key (keys %$conf) {
            next if $key !~ m/^hostpci(\d+)$/;
            my $hostpciindex = $1;


@@ -5465,6 +5617,7 @@ sub vm_stop {
                my $opts = PVE::JSONSchema::pve_parse_startup_order($conf->{startup});
                $timeout = $opts->{down} if $opts->{down};
            }
                my $opts = PVE::JSONSchema::pve_parse_startup_order($conf->{startup});
                $timeout = $opts->{down} if $opts->{down};
            }

+           PVE::GuestHelpers::exec_hookscript($conf, $vmid, 'pre-stop');

        }
 
        $timeout = 60 if !defined($timeout);
        }
 
        $timeout = 60 if !defined($timeout);


@@ -5529,17 +5682,84 @@ sub vm_stop {
 }
 
 sub vm_suspend {
 }
 
 sub vm_suspend {

-    my ($vmid, $skiplock) = @_;
+    my ($vmid, $skiplock, $includestate, $statestorage) = @_;
+
+    my $conf;
+    my $path;
+    my $storecfg;
+    my $vmstate;

 
     PVE::QemuConfig->lock_config($vmid, sub {
 
 
     PVE::QemuConfig->lock_config($vmid, sub {
 

-       my $conf = PVE::QemuConfig->load_config($vmid);
+       $conf = PVE::QemuConfig->load_config($vmid);

 
 

+       my $is_backing_up = PVE::QemuConfig->has_lock($conf, 'backup');

        PVE::QemuConfig->check_lock($conf)
        PVE::QemuConfig->check_lock($conf)

-           if !($skiplock || PVE::QemuConfig->has_lock($conf, 'backup'));
+           if !($skiplock || $is_backing_up);

 
 

-       vm_mon_cmd($vmid, "stop");
+       die "cannot suspend to disk during backup\n"
+           if $is_backing_up && $includestate;
+
+       if ($includestate) {
+           $conf->{lock} = 'suspending';
+           my $date = strftime("%Y-%m-%d", localtime(time()));
+           $storecfg = PVE::Storage::config();
+           $vmstate = PVE::QemuConfig->__snapshot_save_vmstate($vmid, $conf, "suspend-$date", $storecfg, $statestorage, 1);
+           $path = PVE::Storage::path($storecfg, $vmstate);
+           PVE::QemuConfig->write_config($vmid, $conf);
+       } else {
+           vm_mon_cmd($vmid, "stop");
+       }

     });
     });

+
+    if ($includestate) {
+       # save vm state
+       PVE::Storage::activate_volumes($storecfg, [$vmstate]);
+
+       eval {
+           vm_mon_cmd($vmid, "savevm-start", statefile => $path);
+           for(;;) {
+               my $state = vm_mon_cmd_nocheck($vmid, "query-savevm");
+               if (!$state->{status}) {
+                   die "savevm not active\n";
+               } elsif ($state->{status} eq 'active') {
+                   sleep(1);
+                   next;
+               } elsif ($state->{status} eq 'completed') {
+                   print "State saved, quitting\n";
+                   last;
+               } elsif ($state->{status} eq 'failed' && $state->{error}) {
+                   die "query-savevm failed with error '$state->{error}'\n"
+               } else {
+                   die "query-savevm returned status '$state->{status}'\n";
+               }
+           }
+       };
+       my $err = $@;
+
+       PVE::QemuConfig->lock_config($vmid, sub {
+           $conf = PVE::QemuConfig->load_config($vmid);
+           if ($err) {
+               # cleanup, but leave suspending lock, to indicate something went wrong
+               eval {
+                   vm_mon_cmd($vmid, "savevm-end");
+                   PVE::Storage::deactivate_volumes($storecfg, [$vmstate]);
+                   PVE::Storage::vdisk_free($storecfg, $vmstate);
+                   delete $conf->@{qw(vmstate runningmachine)};
+                   PVE::QemuConfig->write_config($vmid, $conf);
+               };
+               warn $@ if $@;
+               die $err;
+           }
+
+           die "lock changed unexpectedly\n"
+               if !PVE::QemuConfig->has_lock($conf, 'suspending');
+
+           vm_qmp_command($vmid, { execute => "quit" });
+           $conf->{lock} = 'suspended';
+           PVE::QemuConfig->write_config($vmid, $conf);
+       });
+    }

 }
 
 sub vm_resume {
 }
 
 sub vm_resume {


@@ -5684,7 +5904,6 @@ sub restore_update_config_line {
     return if $line =~ m/^lock:/;
     return if $line =~ m/^unused\d+:/;
     return if $line =~ m/^parent:/;
     return if $line =~ m/^lock:/;
     return if $line =~ m/^unused\d+:/;
     return if $line =~ m/^parent:/;

-    return if $line =~ m/^template:/; # restored VM is never a template

 
     my $dc = PVE::Cluster::cfs_read_file('datacenter.cfg');
     if (($line =~ m/^(vlan(\d+)):\s*(\S+)\s*$/)) {
 
     my $dc = PVE::Cluster::cfs_read_file('datacenter.cfg');
     if (($line =~ m/^(vlan(\d+)):\s*(\S+)\s*$/)) {


@@ -6367,7 +6586,7 @@ sub do_snapshots_with_qemu {
 
     my $storage_name = PVE::Storage::parse_volume_id($volid);
 
 
     my $storage_name = PVE::Storage::parse_volume_id($volid);
 

-    if ($qemu_snap_storage->{$storecfg->{ids}->{$storage_name}->{type}} 
+    if ($qemu_snap_storage->{$storecfg->{ids}->{$storage_name}->{type}}

        && !$storecfg->{ids}->{$storage_name}->{krbd}){
        return 1;
     }
        && !$storecfg->{ids}->{$storage_name}->{krbd}){
        return 1;
     }


@@ -6411,6 +6630,23 @@ sub template_create {
     });
 }
 
     });
 }
 

+sub convert_iscsi_path {
+    my ($path) = @_;
+
+    if ($path =~ m|^iscsi://([^/]+)/([^/]+)/(.+)$|) {
+       my $portal = $1;
+       my $target = $2;
+       my $lun = $3;
+
+       my $initiator_name = get_initiator_name();
+
+       return "file.driver=iscsi,file.transport=tcp,file.initiator-name=$initiator_name,".
+              "file.portal=$portal,file.target=$target,file.lun=$lun,driver=raw";
+    }
+
+    die "cannot convert iscsi path '$path', unkown format\n";
+}
+

 sub qemu_img_convert {
     my ($src_volid, $dst_volid, $size, $snapname, $is_zero_initialized) = @_;
 
 sub qemu_img_convert {
     my ($src_volid, $dst_volid, $size, $snapname, $is_zero_initialized) = @_;
 


@@ -6431,13 +6667,32 @@ sub qemu_img_convert {
        my $src_path = PVE::Storage::path($storecfg, $src_volid, $snapname);
        my $dst_path = PVE::Storage::path($storecfg, $dst_volid);
 
        my $src_path = PVE::Storage::path($storecfg, $src_volid, $snapname);
        my $dst_path = PVE::Storage::path($storecfg, $dst_volid);
 

+       my $src_is_iscsi = ($src_path =~ m|^iscsi://|);
+       my $dst_is_iscsi = ($dst_path =~ m|^iscsi://|);
+

        my $cmd = [];
        push @$cmd, '/usr/bin/qemu-img', 'convert', '-p', '-n';
        push @$cmd, '-l', "snapshot.name=$snapname" if($snapname && $src_format eq "qcow2");
        push @$cmd, '-t', 'none' if $dst_scfg->{type} eq 'zfspool';
        push @$cmd, '-T', 'none' if $src_scfg->{type} eq 'zfspool';
        my $cmd = [];
        push @$cmd, '/usr/bin/qemu-img', 'convert', '-p', '-n';
        push @$cmd, '-l', "snapshot.name=$snapname" if($snapname && $src_format eq "qcow2");
        push @$cmd, '-t', 'none' if $dst_scfg->{type} eq 'zfspool';
        push @$cmd, '-T', 'none' if $src_scfg->{type} eq 'zfspool';

-       push @$cmd, '-f', $src_format, '-O', $dst_format, $src_path;
-       if ($is_zero_initialized) {
+
+       if ($src_is_iscsi) {
+           push @$cmd, '--image-opts';
+           $src_path = convert_iscsi_path($src_path);
+       } else {
+           push @$cmd, '-f', $src_format;
+       }
+
+       if ($dst_is_iscsi) {
+           push @$cmd, '--target-image-opts';
+           $dst_path = convert_iscsi_path($dst_path);
+       } else {
+           push @$cmd, '-O', $dst_format;
+       }
+
+       push @$cmd, $src_path;
+
+       if (!$dst_is_iscsi && $is_zero_initialized) {

            push @$cmd, "zeroinit:$dst_path";
        } else {
            push @$cmd, $dst_path;
            push @$cmd, "zeroinit:$dst_path";
        } else {
            push @$cmd, $dst_path;


@@ -6472,7 +6727,7 @@ sub qemu_img_format {
 }
 
 sub qemu_drive_mirror {
 }
 
 sub qemu_drive_mirror {

-    my ($vmid, $drive, $dst_volid, $vmiddst, $is_zero_initialized, $jobs, $skipcomplete, $qga) = @_;
+    my ($vmid, $drive, $dst_volid, $vmiddst, $is_zero_initialized, $jobs, $skipcomplete, $qga, $bwlimit) = @_;

 
     $jobs = {} if !$jobs;
 
 
     $jobs = {} if !$jobs;
 


@@ -6499,13 +6754,19 @@ sub qemu_drive_mirror {
     my $opts = { timeout => 10, device => "drive-$drive", mode => "existing", sync => "full", target => $qemu_target };
     $opts->{format} = $format if $format;
 
     my $opts = { timeout => 10, device => "drive-$drive", mode => "existing", sync => "full", target => $qemu_target };
     $opts->{format} = $format if $format;
 

-    print "drive mirror is starting for drive-$drive\n";
-
-    eval { vm_mon_cmd($vmid, "drive-mirror", %$opts); }; #if a job already run for this device,it's throw an error
+    if (defined($bwlimit)) {
+       $opts->{speed} = $bwlimit * 1024;
+       print "drive mirror is starting for drive-$drive with bandwidth limit: ${bwlimit} KB/s\n";
+    } else {
+       print "drive mirror is starting for drive-$drive\n";
+    }

 
 

+    # if a job already runs for this device we get an error, catch it for cleanup
+    eval { vm_mon_cmd($vmid, "drive-mirror", %$opts); };

     if (my $err = $@) {
        eval { PVE::QemuServer::qemu_blockjobs_cancel($vmid, $jobs) };
     if (my $err = $@) {
        eval { PVE::QemuServer::qemu_blockjobs_cancel($vmid, $jobs) };

-       die "mirroring error: $err";
+       warn "$@\n" if $@;
+       die "mirroring error: $err\n";

     }
 
     qemu_drive_mirror_monitor ($vmid, $vmiddst, $jobs, $skipcomplete, $qga);
     }
 
     qemu_drive_mirror_monitor ($vmid, $vmiddst, $jobs, $skipcomplete, $qga);


@@ -6643,7 +6904,7 @@ sub qemu_blockjobs_cancel {
 
 sub clone_disk {
     my ($storecfg, $vmid, $running, $drivename, $drive, $snapname,
 
 sub clone_disk {
     my ($storecfg, $vmid, $running, $drivename, $drive, $snapname,

-       $newvmid, $storage, $format, $full, $newvollist, $jobs, $skipcomplete, $qga) = @_;
+       $newvmid, $storage, $format, $full, $newvollist, $jobs, $skipcomplete, $qga, $bwlimit) = @_;

 
     my $newvolid;
 
 
     my $newvolid;
 


@@ -6678,6 +6939,7 @@ sub clone_disk {
 
        my $sparseinit = PVE::Storage::volume_has_feature($storecfg, 'sparseinit', $newvolid);
        if (!$running || $snapname) {
 
        my $sparseinit = PVE::Storage::volume_has_feature($storecfg, 'sparseinit', $newvolid);
        if (!$running || $snapname) {

+           # TODO: handle bwlimits

            qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit);
        } else {
 
            qemu_img_convert($drive->{file}, $newvolid, $size, $snapname, $sparseinit);
        } else {
 


@@ -6687,7 +6949,7 @@ sub clone_disk {
                    if $drive->{iothread};
            }
 
                    if $drive->{iothread};
            }
 

-           qemu_drive_mirror($vmid, $drivename, $newvolid, $newvmid, $sparseinit, $jobs, $skipcomplete, $qga);
+           qemu_drive_mirror($vmid, $drivename, $newvolid, $newvmid, $sparseinit, $jobs, $skipcomplete, $qga, $bwlimit);

        }
     }
 
        }
     }
 


@@ -6833,12 +7095,15 @@ sub scsihw_infos {
 }
 
 sub add_hyperv_enlightenments {
 }
 
 sub add_hyperv_enlightenments {

-    my ($cpuFlags, $winversion, $machine_type, $kvmver, $bios, $gpu_passthrough) = @_;
+    my ($cpuFlags, $winversion, $machine_type, $kvmver, $bios, $gpu_passthrough, $hv_vendor_id) = @_;

 
     return if $winversion < 6;
     return if $bios && $bios eq 'ovmf' && $winversion < 8;
 
 
     return if $winversion < 6;
     return if $bios && $bios eq 'ovmf' && $winversion < 8;
 

-    push @$cpuFlags , 'hv_vendor_id=proxmox' if $gpu_passthrough;
+    if ($gpu_passthrough || defined($hv_vendor_id)) {
+       $hv_vendor_id //= 'proxmox';
+       push @$cpuFlags , "hv_vendor_id=$hv_vendor_id";
+    }

 
     if (qemu_machine_feature_enabled ($machine_type, $kvmver, 2, 3)) {
        push @$cpuFlags , 'hv_spinlocks=0x1fff';
 
     if (qemu_machine_feature_enabled ($machine_type, $kvmver, 2, 3)) {
        push @$cpuFlags , 'hv_spinlocks=0x1fff';