diff --git a/src/scripts/enable_passthrough_for_pcie_nic_to_kvm_vm.pl b/src/scripts/enable_passthrough_for_pcie_nic_to_kvm_vm.pl
new file mode 100644
index 00000000..c69c614f
--- /dev/null
+++ b/src/scripts/enable_passthrough_for_pcie_nic_to_kvm_vm.pl
@@ -0,0 +1,109 @@
+#!/usr/bin/perl
+
+use strict;
+use warnings;
+
+use Data::Dumper;
+
+my $vm_name = "fastnetmonvm.fastvps.ru";
+
+execute_precheck();
+execute_detach();
+
+sub execute_precheck {
+    my $cmdline = `cat /proc/cmdline`;
+    chomp $cmdline;
+ 
+    unless ($cmdline =~ /intel_iommu=on/) {
+        print "Please add intel_iommu=on to kernel params\n";
+        
+        print "You could do it in file /etc/default/grub\n";
+        print 'With param: GRUB_CMDLINE_LINUX_DEFAULT="intel_iommu=on"', "\n";
+        print "update-grub\n";
+        print "reboot\n";
+
+        exit(1);
+    }  
+    
+    my $conf_path = '/etc/modprobe.d/vfio_iommu_type1.conf';
+
+    # Debian Jessie, 3.16, Intel(R) Core(TM) i7-3820 CPU @ 3.60GHz desktop 
+    # Could be fixed in runtime:
+    # echo 1 > /sys/module/vfio_iommu_type1/parameters/allow_unsafe_interrupts
+    unless (-e $conf_path) {
+        print "Please apply work around for error\n";
+        print "vfio_iommu_type1_attach_group: No interrupt remapping support.\n\n";
+
+        print "echo \"options vfio_iommu_type1 allow_unsafe_interrupts=1\" > /etc/modprobe.d/vfio_iommu_type1.conf";
+        print "\n";
+        print "And reboot server \n";
+        
+        exit(0);
+    }
+}
+
+sub execute_detach {
+    # 03:00.0 Ethernet controller: Intel Corporation 82599ES 10-Gigabit SFI/SFP+ Network Connection (rev 01)
+    my @lspci = `lspci`;
+    chomp @lspci;
+
+    # We process only Ethernet devices
+    @lspci = grep {/Ethernet/} @lspci;
+
+    @lspci = grep {/82599/} @lspci;
+
+    my @nic_addresses = ();
+
+    for my $nic (@lspci) {
+        if ($nic =~ /(\d+\:\d+\.\d+)/) {
+            push @nic_addresses, $1;
+        }
+    }
+
+    my @virsh_full_addresses = ();
+    for my $nic_address (@nic_addresses) {
+        my $nic_address_in_virsh_format = $nic_address;
+        $nic_address_in_virsh_format =~ s/[\:\.]/_/;
+
+        my $virsh_address_full_format = `virsh nodedev-list | grep '$nic_address_in_virsh_format'`;
+        chomp $virsh_address_full_format;
+
+        push @virsh_full_addresses, $virsh_address_full_format;
+    }
+
+    # We use hash because multy port NICs could produce multiple equal address groups
+    my $xml_blocks = {};
+
+    print "Detach NICs from the system\n";
+    for my $virsh_address (@virsh_full_addresses) {
+        my $output = `virsh nodedev-dettach $virsh_address 2>&1`;
+        chomp $output;
+
+        if ($? != 0) {
+            die "virsh nodedev-dettach failed with output: $output\n";
+        }
+   
+        # <address domain='0x0000' bus='0x03' slot='0x00' function='0x0'/> 
+        my @xml_address_data = `virsh nodedev-dumpxml $virsh_address | grep address`;
+        chomp @xml_address_data;
+
+        for my $xml_line (@xml_address_data) {
+            # cleanup
+            $xml_line =~ s/^\s+//g;
+            $xml_line =~ s/\s+$//g;
+
+            $xml_blocks->{ $xml_line } = 1;
+        }
+    }
+
+    my $target_xml = '';
+    for my $address_block (keys %$xml_blocks) {
+        $target_xml .= "<hostdev mode='subsystem' type='pci' managed='yes'><source>$address_block</source></hostdev>\n";
+    }
+
+    print "Please run virsh edit $vm_name and insert this xml to devices block\n\n";
+    
+    print $target_xml, "\n";
+
+    print "After this please execute virsh destroy $vm_name and virsh start $vm_name for applying changes\n";
+}